This document provides an explanation of when to use a repeated measures ANOVA. It describes how a researcher would make decisions to determine if a repeated measures ANOVA is the appropriate statistical test. Key factors include having one dependent variable, one within-subjects independent variable with three or more levels, repeated measures on the same subjects over time, and the goal of testing for differences in mean scores across time periods. An example is provided of administering the same exam to students over three time periods to see if their skills improve over the semester.
Reporting Pearson Correlation Test of Independence in APAKen Plummer
A Pearson correlation test of independence was conducted to determine if student height and GPA were related. A weak correlation was found between height and GPA (r = .217, p > .05), indicating that student height and GPA are independent of each other.
Reporting single sample z-test for proportionsKen Plummer
A friend claimed that 25% of statistics students would use statistics in their future careers, but the author believed it was less. To test this, the author surveyed 200 fellow students (out of 250 asked), and found that only 18% said they would use statistics. The author conducted a z-test for proportions and found a statistically significant difference between the claim of 25% and the sample proportion of 18%, with z=2.35 and p=0.001.
Structural equation modeling (SEM) is used to analyze relationships between multiple independent and dependent variables. It allows for simultaneous testing of these relationships while accounting for measurement error. The goal of SEM is to determine if the estimated population covariance matrix from the model fits the sample covariance matrix. It can be used to test theories, account for variance, and assess reliability and parameter estimates. Key considerations include sample size, normality, linearity, and identification of the model. Model fit is assessed using absolute, comparative, and parsimonious fit indices. Modification indices can also indicate how to improve model fit.
The document provides guidance on reporting the results of a single sample t-test in APA format. It includes an example result that states there was no statistically significant difference in calculus anxiety scores between a sample of 30 students and the general college student population based on a t-value of 1.03 and p-value of 0.434. Key elements to report include the sample mean and standard deviation, degrees of freedom, t-value, and p-value.
This presentation discusses parametric and non-parametric methods for analyzing relationships between variables. Parametric methods can be used when sample data is normally distributed and scaled, representing population parameters. They involve examining relationships between variables like death anxiety and religiosity through statistical tests. Non-parametric methods do not require normal distribution or scaling and can be used as an alternative.
This document provides guidance on reporting the results of a single sample t-test in APA format. It includes templates for describing the test and population in the introduction and reporting the mean, standard deviation, t-value and significance in the results. An example is given of a hypothetical single sample t-test comparing IQ scores of people who eat broccoli regularly to the general population.
Towards Explainable Fact Checking (DIKU Business Club presentation)Isabelle Augenstein
Outline:
- Fact checking – what is it and why do we need it?
- False information online
- Content-based automatic fact checking
- Explainability – what is it and why do we need it?
- Making the right predictions for the right reasons
- Model training pipeline
- Explainable fact checking – some first solutions
- Rationale selection
- Generating free-text explanations
- Wrap-up
Reporting Pearson Correlation Test of Independence in APAKen Plummer
A Pearson correlation test of independence was conducted to determine if student height and GPA were related. A weak correlation was found between height and GPA (r = .217, p > .05), indicating that student height and GPA are independent of each other.
Reporting single sample z-test for proportionsKen Plummer
A friend claimed that 25% of statistics students would use statistics in their future careers, but the author believed it was less. To test this, the author surveyed 200 fellow students (out of 250 asked), and found that only 18% said they would use statistics. The author conducted a z-test for proportions and found a statistically significant difference between the claim of 25% and the sample proportion of 18%, with z=2.35 and p=0.001.
Structural equation modeling (SEM) is used to analyze relationships between multiple independent and dependent variables. It allows for simultaneous testing of these relationships while accounting for measurement error. The goal of SEM is to determine if the estimated population covariance matrix from the model fits the sample covariance matrix. It can be used to test theories, account for variance, and assess reliability and parameter estimates. Key considerations include sample size, normality, linearity, and identification of the model. Model fit is assessed using absolute, comparative, and parsimonious fit indices. Modification indices can also indicate how to improve model fit.
The document provides guidance on reporting the results of a single sample t-test in APA format. It includes an example result that states there was no statistically significant difference in calculus anxiety scores between a sample of 30 students and the general college student population based on a t-value of 1.03 and p-value of 0.434. Key elements to report include the sample mean and standard deviation, degrees of freedom, t-value, and p-value.
This presentation discusses parametric and non-parametric methods for analyzing relationships between variables. Parametric methods can be used when sample data is normally distributed and scaled, representing population parameters. They involve examining relationships between variables like death anxiety and religiosity through statistical tests. Non-parametric methods do not require normal distribution or scaling and can be used as an alternative.
This document provides guidance on reporting the results of a single sample t-test in APA format. It includes templates for describing the test and population in the introduction and reporting the mean, standard deviation, t-value and significance in the results. An example is given of a hypothetical single sample t-test comparing IQ scores of people who eat broccoli regularly to the general population.
Towards Explainable Fact Checking (DIKU Business Club presentation)Isabelle Augenstein
Outline:
- Fact checking – what is it and why do we need it?
- False information online
- Content-based automatic fact checking
- Explainability – what is it and why do we need it?
- Making the right predictions for the right reasons
- Model training pipeline
- Explainable fact checking – some first solutions
- Rationale selection
- Generating free-text explanations
- Wrap-up
The document presents a case study for a lead scoring model built to predict potential customer conversions for an education company. Data on past leads was analyzed to identify key variables impacting conversion rate. A logistic regression model was developed and evaluated on train and test data, achieving 78% accuracy. The model can assign a lead score between 0-100 to help the company prioritize hot leads most likely to convert.
This document discusses regression models, path models, and the output from AMOS software when conducting structural equation modeling (SEM). Regression models only include observed variables and assume independents are measured without error. Path models allow independents to be both causes and effects, and allow for error terms on endogenous variables. The AMOS output provides standardized and unstandardized regression weights, significance tests, and fit indexes to evaluate how well the specified model fits the sample data.
This document discusses effect size, which is a measure used to quantify the size of the difference between two groups. It is calculated by taking the difference between the means of two groups and dividing by the standard deviation. Effect sizes of 0.2, 0.5, and 0.8 are considered small, medium, and large, respectively. Meta-analyses often use effect sizes to combine results from different studies by standardizing outcomes. Visible Learning, a meta-analysis of 800 other meta-analyses, found that most educational interventions have a small positive effect size of around 0.4, which corresponds to a two year increase in achievement. Effect sizes are useful for comparing impacts across different studies and determining what types of teaching practices have
This document discusses how to report the results of a Pearson correlation analysis in APA style. It provides an example of a problem investigating the relationship between broccoli extract consumption and well-being scores. The template shown reports that a strong positive correlation was found between broccoli extract consumption and well-being (r = .88, p < .05).
Reporting chi square goodness of fit test of independence in apaKen Plummer
A chi-square goodness of fit test was used to analyze data from a public opinion poll of 1000 voters in Connecticut on their party affiliation. The expected distribution was 40% Republican and 60% Democrat, but the observed results were 32% Republican and 68% Democrat. A sample report in APA style for these results includes the chi-square value, degrees of freedom, and p-value to determine if there is a significant deviation from the expected distribution.
- Regression analysis is a statistical technique used to measure the relationship between two quantitative variables and make causal inferences.
- A regression model graphs the relationship between a dependent variable (Y axis) and one or more independent variables (X axis). The goal is to find the linear equation that best fits the data.
- The regression equation takes the form Y = a + bX, where a is the intercept, b is the slope coefficient, and X and Y are the variables. The coefficient b indicates the strength and direction of the relationship.
Simulating data to gain insights intopower and p-hackingDorothy Bishop
Very basic introduction to simulating data to illustrate issues affecting reproducibility. Uses Excel and R, but assumes no prior knowledge of R. Please let me know of errors or things that need better explanation.
This document provides an overview of single linear regression. It explains that single linear regression extends the concept of correlation by using one variable to predict the value of another variable. It discusses using scatter plots to visualize the relationship between two variables and determine if the relationship is strong or weak, and whether it is positive or negative. Examples are provided to illustrate single linear regression concepts and how to interpret different types of relationships between variables.
Slides from my PhD defense
Example-Dependent Cost-Sensitive Classification
Applications in Financial Risk Modeling and Marketing Analytics
https://github.com/albahnsen/phd-thesis
Statistics - Simple Linear and Multiple Linear RegressionBryll Edison Par
Introduction to simple and multiple linear regression.
https://issuu.com/arbrylledisonparmodules/docs/archi203_par_report_multiple_and_simple_linear_reg
Null hypothesis for partial correlationKen Plummer
The document discusses setting a null hypothesis for a partial correlation. It provides a template for a null hypothesis when testing the relationship between two variables while adjusting for a third variable. As an example, it gives the null hypothesis that there is no relationship between plant growth and certain amounts of fertilizer, adjusting for sunlight.
Regression (Linear Regression and Logistic Regression) by Akanksha BaliAkanksha Bali
Regression analysis is a statistical technique used to examine relationships between variables. Linear regression finds the best fitting straight line through data points to model the relationship between a continuous dependent variable (Y) and one or more independent variables (X). Logistic regression produces results in a binary format to predict outcomes of categorical dependent variables. It transforms the linear equation using logarithms to restrict predicted Y values between 0 and 1.
1. Researchers should consult multiple fit statistics when evaluating the fit of a confirmatory factor analysis model as no single statistic is ideal.
2. Different fit statistics were developed with different rationales and assess model fit in various ways.
3. Sample size impacts the chi-square statistic, with larger samples increasing the likelihood of rejection.
The document discusses different types of two-sample hypothesis tests, including tests comparing two population means of independent samples, two population proportions, and paired or dependent samples. It provides examples and step-by-step explanations of how to conduct two-sample t-tests, z-tests, and tests of proportions. Key points covered include determining the appropriate test statistic based on sample size and characteristics, stating the null and alternative hypotheses, test criteria, and decisions rules.
This document discusses building a multiple linear regression model to predict profit using backward elimination. It imports datasets, encodes categorical variables, splits data into training and test sets, fits a regression model to the training set, predicts results on the test set, and builds an optimal model through iterative backward elimination of insignificant variables. Key steps include encoding state as a factor, sequentially removing variables with high p-values from the model, and evaluating the models.
The presentation covered key steps in analyzing survey data including defining goals, designing valid and reliable survey questions, collecting data, cleaning data, conducting descriptive statistics and correlations, comparing mean differences between groups, and clearly presenting results along with conclusions and recommendations. Piloting surveys and continuously improving methods was also emphasized.
Ibmathstudiesinternalassessmentfinaldraft 101208070253-phpapp02Travis Hayes
This document is a math studies internal assessment that investigates the relationship between SAT scores and family income of test takers around the world. The student analyzed data on SAT scores and family incomes from the College Board in 2007. A scatter plot, least squares regression line, and correlation coefficient calculation showed a strong positive correlation between higher SAT scores and higher family incomes. A chi-squared test rejected the null hypothesis that SAT scores and family income are independent. However, limitations in the data are noted, such as incomplete income reporting and wide income brackets in the raw data.
How to input data in spss for independent samples t testsKen Plummer
The document discusses how to input data into SPSS to conduct an independent samples t-test to determine if there is a difference in ACT scores between males and females. The variables - gender (independent variable) and ACT scores (dependent variable) - need to be named and coded as nominal/scale data respectively. Numeric values of 1 and 2 are assigned to males and females. The raw data consisting of gender codes and ACT scores is then input into the SPSS data view to allow analysis and interpretation of results.
The document discusses shared variance and effect size. It explains that shared variance, as measured by r-squared, indicates the percentage of variability between two variables that is explained by their relationship. Effect size can be measured by Cohen's d or eta-squared and indicates the size of the difference or amount of variability explained by a treatment group. Calculating measures of shared variance and effect size provides information about the strength and importance of relationships in statistical analyses beyond just significance.
Reporting a one way repeated measures anovaKen Plummer
The document provides guidance on reporting the results of a one-way repeated measures ANOVA in APA style. It includes templates for reporting the main ANOVA results and any post-hoc pairwise comparisons between conditions. Key sections are highlighted to fill in values from an example SPSS output to generate a complete APA-style results section reporting a significant effect of time of season on pizza consumption.
This document discusses the null hypothesis for a one-way analysis of covariance (ANCOVA). It explains that a one-way ANCOVA compares the influence of an independent variable with at least two levels on a dependent variable, while controlling for the effect of a covariate. The document provides a template for writing the null hypothesis, which states that there is no significant effect of the independent variable on the dependent variable when controlling for the covariate. It gives two examples applying this template.
The document provides guidance on reporting the results of an ANCOVA analysis in APA format. It recommends including that a one-way ANCOVA was conducted to determine differences between levels of an independent variable on a dependent variable while controlling for a covariate. An example is given using athlete type as the independent variable, slices of pizza eaten as the dependent variable, and weight as the covariate. The document also provides a template for reporting the F-ratio, degrees of freedom, and significance level.
The document presents a case study for a lead scoring model built to predict potential customer conversions for an education company. Data on past leads was analyzed to identify key variables impacting conversion rate. A logistic regression model was developed and evaluated on train and test data, achieving 78% accuracy. The model can assign a lead score between 0-100 to help the company prioritize hot leads most likely to convert.
This document discusses regression models, path models, and the output from AMOS software when conducting structural equation modeling (SEM). Regression models only include observed variables and assume independents are measured without error. Path models allow independents to be both causes and effects, and allow for error terms on endogenous variables. The AMOS output provides standardized and unstandardized regression weights, significance tests, and fit indexes to evaluate how well the specified model fits the sample data.
This document discusses effect size, which is a measure used to quantify the size of the difference between two groups. It is calculated by taking the difference between the means of two groups and dividing by the standard deviation. Effect sizes of 0.2, 0.5, and 0.8 are considered small, medium, and large, respectively. Meta-analyses often use effect sizes to combine results from different studies by standardizing outcomes. Visible Learning, a meta-analysis of 800 other meta-analyses, found that most educational interventions have a small positive effect size of around 0.4, which corresponds to a two year increase in achievement. Effect sizes are useful for comparing impacts across different studies and determining what types of teaching practices have
This document discusses how to report the results of a Pearson correlation analysis in APA style. It provides an example of a problem investigating the relationship between broccoli extract consumption and well-being scores. The template shown reports that a strong positive correlation was found between broccoli extract consumption and well-being (r = .88, p < .05).
Reporting chi square goodness of fit test of independence in apaKen Plummer
A chi-square goodness of fit test was used to analyze data from a public opinion poll of 1000 voters in Connecticut on their party affiliation. The expected distribution was 40% Republican and 60% Democrat, but the observed results were 32% Republican and 68% Democrat. A sample report in APA style for these results includes the chi-square value, degrees of freedom, and p-value to determine if there is a significant deviation from the expected distribution.
- Regression analysis is a statistical technique used to measure the relationship between two quantitative variables and make causal inferences.
- A regression model graphs the relationship between a dependent variable (Y axis) and one or more independent variables (X axis). The goal is to find the linear equation that best fits the data.
- The regression equation takes the form Y = a + bX, where a is the intercept, b is the slope coefficient, and X and Y are the variables. The coefficient b indicates the strength and direction of the relationship.
Simulating data to gain insights intopower and p-hackingDorothy Bishop
Very basic introduction to simulating data to illustrate issues affecting reproducibility. Uses Excel and R, but assumes no prior knowledge of R. Please let me know of errors or things that need better explanation.
This document provides an overview of single linear regression. It explains that single linear regression extends the concept of correlation by using one variable to predict the value of another variable. It discusses using scatter plots to visualize the relationship between two variables and determine if the relationship is strong or weak, and whether it is positive or negative. Examples are provided to illustrate single linear regression concepts and how to interpret different types of relationships between variables.
Slides from my PhD defense
Example-Dependent Cost-Sensitive Classification
Applications in Financial Risk Modeling and Marketing Analytics
https://github.com/albahnsen/phd-thesis
Statistics - Simple Linear and Multiple Linear RegressionBryll Edison Par
Introduction to simple and multiple linear regression.
https://issuu.com/arbrylledisonparmodules/docs/archi203_par_report_multiple_and_simple_linear_reg
Null hypothesis for partial correlationKen Plummer
The document discusses setting a null hypothesis for a partial correlation. It provides a template for a null hypothesis when testing the relationship between two variables while adjusting for a third variable. As an example, it gives the null hypothesis that there is no relationship between plant growth and certain amounts of fertilizer, adjusting for sunlight.
Regression (Linear Regression and Logistic Regression) by Akanksha BaliAkanksha Bali
Regression analysis is a statistical technique used to examine relationships between variables. Linear regression finds the best fitting straight line through data points to model the relationship between a continuous dependent variable (Y) and one or more independent variables (X). Logistic regression produces results in a binary format to predict outcomes of categorical dependent variables. It transforms the linear equation using logarithms to restrict predicted Y values between 0 and 1.
1. Researchers should consult multiple fit statistics when evaluating the fit of a confirmatory factor analysis model as no single statistic is ideal.
2. Different fit statistics were developed with different rationales and assess model fit in various ways.
3. Sample size impacts the chi-square statistic, with larger samples increasing the likelihood of rejection.
The document discusses different types of two-sample hypothesis tests, including tests comparing two population means of independent samples, two population proportions, and paired or dependent samples. It provides examples and step-by-step explanations of how to conduct two-sample t-tests, z-tests, and tests of proportions. Key points covered include determining the appropriate test statistic based on sample size and characteristics, stating the null and alternative hypotheses, test criteria, and decisions rules.
This document discusses building a multiple linear regression model to predict profit using backward elimination. It imports datasets, encodes categorical variables, splits data into training and test sets, fits a regression model to the training set, predicts results on the test set, and builds an optimal model through iterative backward elimination of insignificant variables. Key steps include encoding state as a factor, sequentially removing variables with high p-values from the model, and evaluating the models.
The presentation covered key steps in analyzing survey data including defining goals, designing valid and reliable survey questions, collecting data, cleaning data, conducting descriptive statistics and correlations, comparing mean differences between groups, and clearly presenting results along with conclusions and recommendations. Piloting surveys and continuously improving methods was also emphasized.
Ibmathstudiesinternalassessmentfinaldraft 101208070253-phpapp02Travis Hayes
This document is a math studies internal assessment that investigates the relationship between SAT scores and family income of test takers around the world. The student analyzed data on SAT scores and family incomes from the College Board in 2007. A scatter plot, least squares regression line, and correlation coefficient calculation showed a strong positive correlation between higher SAT scores and higher family incomes. A chi-squared test rejected the null hypothesis that SAT scores and family income are independent. However, limitations in the data are noted, such as incomplete income reporting and wide income brackets in the raw data.
How to input data in spss for independent samples t testsKen Plummer
The document discusses how to input data into SPSS to conduct an independent samples t-test to determine if there is a difference in ACT scores between males and females. The variables - gender (independent variable) and ACT scores (dependent variable) - need to be named and coded as nominal/scale data respectively. Numeric values of 1 and 2 are assigned to males and females. The raw data consisting of gender codes and ACT scores is then input into the SPSS data view to allow analysis and interpretation of results.
The document discusses shared variance and effect size. It explains that shared variance, as measured by r-squared, indicates the percentage of variability between two variables that is explained by their relationship. Effect size can be measured by Cohen's d or eta-squared and indicates the size of the difference or amount of variability explained by a treatment group. Calculating measures of shared variance and effect size provides information about the strength and importance of relationships in statistical analyses beyond just significance.
Reporting a one way repeated measures anovaKen Plummer
The document provides guidance on reporting the results of a one-way repeated measures ANOVA in APA style. It includes templates for reporting the main ANOVA results and any post-hoc pairwise comparisons between conditions. Key sections are highlighted to fill in values from an example SPSS output to generate a complete APA-style results section reporting a significant effect of time of season on pizza consumption.
This document discusses the null hypothesis for a one-way analysis of covariance (ANCOVA). It explains that a one-way ANCOVA compares the influence of an independent variable with at least two levels on a dependent variable, while controlling for the effect of a covariate. The document provides a template for writing the null hypothesis, which states that there is no significant effect of the independent variable on the dependent variable when controlling for the covariate. It gives two examples applying this template.
The document provides guidance on reporting the results of an ANCOVA analysis in APA format. It recommends including that a one-way ANCOVA was conducted to determine differences between levels of an independent variable on a dependent variable while controlling for a covariate. An example is given using athlete type as the independent variable, slices of pizza eaten as the dependent variable, and weight as the covariate. The document also provides a template for reporting the F-ratio, degrees of freedom, and significance level.
This document provides an overview of analysis of variance (ANOVA) techniques. It explains that ANOVA allows comparison of three or more group means and extends the independent t-test. The null hypothesis is that all group means are equal, while the alternative is that at least one pair of means is different. Key assumptions are listed. Steps for computing ANOVA are outlined, including calculating sums of squares, degrees of freedom, mean squares, and the F statistic. An example compares sales under different offer conditions to illustrate the computation and interpretation of ANOVA results.
1) ANOVA is used to compare the means of more than two populations and determine if observed differences are due to chance or actual differences in the population means.
2) The document provides an example of using a one-way single factor ANOVA to analyze the effects of different teaching formats on student exam scores.
3) The ANOVA compares the between-treatment variability to the within-treatment variability using an F-test. If the between-treatment variability is significantly larger, it suggests the population means differ. In this example, the F-test showed no significant difference between the teaching formats.
This document discusses three applications of one-way ANOVA:
1) It analyzes summer earnings data for B.A., B.Sc., and B.B.A. students and finds a statistically significant difference between programs.
2) It examines data on three margarine brands and finds no sufficient evidence of differences between brands.
3) Job offer data for MBA majors in finance, marketing, and management show statistically significant differences between the majors.
This document discusses using two-way ANOVA to analyze two different experiments. The first experiment tests for differences in headache improvement between four drug mixtures and two injection schedules. The two-way ANOVA results show there are differences between the drug mixtures but not the schedules, and an interaction between the two factors. The second experiment analyzes weight loss between four diets in five age groups. The two-way ANOVA results show no significant differences between the four diets at the 1% level.
This document provides information about analysis of variance (ANOVA). It discusses different types of ANOVA based on the number of independent variables and samples. A one-way ANOVA has one independent variable and can be independent or repeated measures. Independent ANOVA involves different samples for each group, while repeated measures ANOVA involves measuring the same sample under different conditions. The document provides an example of a one-way repeated measures ANOVA comparing student test scores on different days to determine if their knowledge differed significantly between days. It outlines transferring the data to SPSS and running appropriate tests such as Mauchly's test of sphericity and Bonferroni post hoc test for repeated measures ANOVA.
ANOVA is a statistical technique used to determine whether the means of groups are statistically different from each other. It can be used to establish cause-and-effect relationships with a certain degree of certainty. There are different types of ANOVA for different study designs. The basic parts of an ANOVA include sums of squares, degrees of freedom, mean squares, and the F-statistic. ANOVA can be performed in Excel using the data analysis tool. An example shows how ANOVA was used to analyze measurement data from multiple inspectors.
The document discusses a one-way ANOVA test, which compares the means of two or more independent groups on a continuous dependent variable. It outlines the assumptions of the test, how to set it up in SPSS, and how to interpret the output. Key outputs include an ANOVA table showing if group means are statistically significantly different, and a post-hoc test for determining the nature of differences between specific groups.
This document provides an overview of analysis of variance (ANOVA). It introduces ANOVA and its key concepts, including its development by Ronald Fisher. It defines ANOVA and distinguishes between one-way and two-way ANOVA. It outlines the assumptions, techniques, and examples of how to perform one-way and two-way ANOVA. It also discusses the uses, advantages, and limitations of ANOVA for analyzing differences between multiple means and factors.
A one-way independent measures ANOVA was used to analyze the effect of warning light color (red, yellow, green, blue) on reaction time. The ANOVA revealed a significant effect of color on reaction time. Post-hoc comparisons using Tukey's test showed participants reacted significantly slower to the blue light compared to the red, yellow, and green lights.
The document describes a study that examined the effects of four different diets on weight gain in rats. Twenty-four rats were given one of four diets that varied in vitamin and protein content over two weeks. The weights were recorded and analyzed using one-way ANOVA and post hoc tests. The analysis found that diet 1 (0.1% vitamin, 10% protein) resulted in significantly greater weight gain than diets 2-4. Diets 2, 3 and 4 were not significantly different from each other in terms of weight gain. Therefore, diet 1 was determined to be the optimal diet for promoting weight gain in rats based on the statistical analysis.
The document discusses levels in statistics and provides examples to illustrate the concept. Levels refer to the number of conditions within an independent variable. The number of levels determines the appropriate statistical analysis method. Examples are provided of studies with different numbers of levels, such as socioeconomic status having 4 levels (wealthy, upper middle class, lower middle class, below poverty line) while gender has 2 levels (male, female). Visual representations are given to depict levels within independent variables. The document concludes by restating that levels indicate the number of conditions in an independent variable and that determining the number of levels is important for selecting the correct statistical analysis.
The document defines the mode as the most frequently occurring observation in a data set. It uses an example data set of [1, 2, 2, 2, 2, 3] to show that the mode is 2, as it occurs most frequently. The mode provides a measure of central tendency but can reveal multiple peaks in a distribution that may not be evident from just the mean or median.
What is a Kendall's Tau (independence)?Ken Plummer
Kendall's Tau is a nonparametric correlation test used with ordinal or ranked data, like ranks in an Ironman competition. It measures the association between two ranked variables, like biking and running event ranks. Researchers would use Kendall's Tau to analyze the degree of dependence between athletes' biking and running ranks in an Ironman event due to ties in ranks. Kendall's Tau results range from -1 to 1, with values closer to the extremes indicating a stronger monotonic relationship and values closer to 0 indicating independence between the variables.
Teach me ordinal or scaled (central tendency)Ken Plummer
Ordinal data ranks items but with unequal intervals between ranks, so the median and inter-quartile range are used to analyze central tendency and spread. Scaled data has equal intervals between units like measurements, so the mean and standard deviation are appropriate measures of central tendency and spread. Both types of data require different statistical analysis approaches depending on whether the data is ordinal or scaled.
Null hypothesis for a single-sample t-testKen Plummer
The document discusses the null hypothesis for a single-sample t-test. It explains that a null hypothesis sets up the probability that there is no effect or relationship, and evidence is then collected to either accept or reject that hypothesis. For a single-sample t-test, the null hypothesis is that there is no significant difference between the mean of the sample being tested and the population from which it was drawn.
The document discusses covariates and how to determine if they are present in a problem. It uses an example of examining the relationship between socioeconomic status (SES) and student ACT scores, and whether parental status is a covariate. Adding parental status as a variable decreases the correlation between SES and ACT scores, showing it plays a role in explaining their relationship. Controlling for covariates allows viewing the unique relationship between variables when the covariate's effect is removed. Key terms that indicate a covariate is present include "control for", "hold constant", and "adjust for".
The document discusses null hypotheses for point-biserial correlations. It explains that a null hypothesis states there is no relationship between a dichotomous variable and a continuous variable. An example null hypothesis is provided: "There is no statistically significant relationship between the student height and college graduation status." The document also provides a template for writing null hypotheses for point-biserial correlations and includes another example.
Quick reminder ordinal, scaled or nominal proportionalKen Plummer
Ordinal data ranks items but with unequal intervals between ranks, like place or percentiles. Scaled data has equal intervals between units, like speed, height, or weight. Nominal proportional data shows proportions or percentages, like percentages in surveys.
A repeated measures ANOVA is used to test whether a single group of people change over time by comparing distributions from the same group at different time periods, rather than comparing distributions from different groups. The overall F-ratio reveals if there are differences among time periods, and post hoc tests identify exactly where the differences occurred. In contrast, a one-way ANOVA compares distributions between two or more different groups to determine if there are statistical differences between them.
The document describes an experiment that aims to test the relationship between an independent variable (IV) and a dependent variable (DV) while controlling for confounding variables. It discusses how the participants were divided into two groups and given different word lists to read aloud. However, because the word lists were either matched or mismatched between the groups, the experiment confounded the manipulated IV with another variable, making it impossible to determine the effect of the IV alone on the DV. The document then discusses various sources of variability in experimental data and different methods of experimental control, such as comparison groups, that can help isolate the effect of the IV.
The document discusses the paired samples t-test, which is used to compare two sets of measurements made on the same individuals. It notes that this test is appropriate when there are two correlated distributions, such as pre-test and post-test scores from the same people. The null hypothesis is that there is no difference between the pairs. The test calculates the differences between pairs, sums them, and divides this by the standard error of the differences to obtain a t-value, which can be compared to critical values to determine if the null hypothesis can be rejected.
This document discusses sampling and experimental control in psychological research. It defines key concepts like population, sample, random selection, and control variables. Random sampling methods like simple random sampling and systematic sampling are described as being better than non-random methods like convenience sampling which can result in biased samples. The importance of controlling extraneous variables through randomization and comparison groups is explained to isolate the effect of independent variables on dependent variables. Common experimental designs are reviewed, highlighting the strengths of designs with random assignment and comparison groups over one-group pre-post designs or designs with non-equivalent groups.
In this presentation, you will differentiate the ANOVA and ANCOVA statistical methods, and identify real-world situations where the ANOVA and ANCOVA methods for statistical inference are applied.
This document outlines the steps taken to determine that a one-way ANOVA is the appropriate statistical analysis for a problem involving comparing English test scores of students based on the native language of their instructor. Specifically, it involves determining that the problem is inferential in nature, examines differences between groups, has a normal distribution, ratio data, one dependent variable, one independent variable with three or more levels, leading to a one-way ANOVA.
This document provides an overview of analysis of variance (ANOVA) tests, including one-way and two-way ANOVA, repeated measures ANOVA, and factorial ANOVA. It explains key concepts like factors, levels, and assumptions. Guidelines are provided for determining what type of ANOVA to use depending on the study design and number of independent and dependent variables. Steps for conducting ANOVA tests and interpreting F-statistics are also outlined. The document compares ANOVA to t-tests and explains why ANOVA is preferable when comparing more than two groups.
Levels of an independent variable (Practice Set)Ken Plummer
The document presents four practice problems involving determining the number of levels of the independent variable in different studies. The first problem involves students requesting to be in the same lunch period as friends or not, which has two levels. The second involves belief in God, which has two levels of presence or absence of belief. The third examines suicidal thoughts across four grade levels. The fourth looks at college acceptance rates between males and females, which has two gender levels.
1 Crosstabs Lesson 1 Running crosstabs to test you.docxhoney725342
1
Crosstabs
Lesson 1: Running crosstabs to test your hypothesis
To access the Crosstabs in your SPSS, click the following:
Analyze Descriptive Statistics Crosstabs
Once you work your way through these selections, you should reach this dialog box:
You will want to place your dependent variable in the row variable and your independent variable in the
column variable. For this example, we will use AFFRMACT (preference for affirmative action policies) for
the dependent variable and SEX (gender of the respondent) as the independent variable.
2
Now we want to tell SPSS to compute the column percentages. To do this you will choose ‘Cells’ and
then select ‘Column’ in the “Percentages” box; then click ‘Continue’.
Now select ‘OK’ to run your crosstabs. You should get the following results:
3
Lesson 2: Examining Your Output
We just ran crosstabs to test a hypothesis with two variables, one nominal (SEX-independent variable)
and one ordinal (AFFRMACT-dependent variable). As you can see, the categories of the independent
variable are found across the top in the columns and the dependent variable information is found down
the side forming the rows. Each square is known as a cell and within each cell is the frequency (or
count) and the column percentage. You can also find the row totals and column totals, which are
sometimes referred to as marginal.
In our example we know the following is true:
7.8% of men in this sample strongly support affirmative action policies, whereas 11.4% of
women do;
We can also look at grouping at a glance and concede that 15% of men and 18.5% of women
support affirmative action policies in comparison to 84.9% of men and 81.5% of women oppose
these policies.
The bottom right cell in the table is where we can find that we had 1,904 people answer this
question as our sample.
Lesson 3: Interpreting Crosstabs
Researchers run crosstabs to determine whether there is an association between two variables. Also,
crosstabs may tell us other important things about the relationship between the two variables, including
the strength of association, and sometimes the direction of the association. FYI, the direction can only
be found when both of the variables in your table are greater than nominal.
Ask yourself the following questions after you populate your crosstabs:
1. Is there an association between the two variables?
If you answer yes to this (or maybe), then move to question 2.
2. What is the strength of association between the two variables?
If BOTH variables are ordinal than move to question 3.
3. What is the direction of association?
Is there an association?
What we are trying to determine here is whether knowing the value of one variable will help us predict
the value of another variable. In other words, if gender is associated with preference to affirmative
action policies. In orde ...
Experimental design cartoon part 5 sample sizeKevin Hamill
Part 5 of 5 - Experimental design lecture series. This one focuses on sample size calculations and introduces some of the commonly used statistical tests (for normally distributed data). Toward the end it covers type I and II errors, alpha/beta and reducing variability.
This document discusses key concepts in experimental design, including types of variables, scales of measurement, and validity and reliability. It describes independent and dependent variables, and how dependent variables are measured and operationalized. It outlines different scales of measurement including nominal, ordinal, interval, and ratio scales. Errors in measurement are discussed, specifically reliability which measures consistency, and validity which measures accuracy. Threats to internal and external validity are presented. The importance of controlling for confounding variables is demonstrated in an example comparing men's and women's performance on a color naming task.
Central tendency, shape, or symmetry practice problems (2)Ken Plummer
The document presents 5 practice problems about differentiating between concepts of central tendency, spread, and symmetry in statistics. For each problem, the user is asked to identify which statistical concept is being examined based on a description. The concepts are then explained, with central tendency referring to average or middle values, spread referring to the differences between lowest and highest values, and symmetry referring to the shape or distribution of data.
The document discusses analysis of variance (ANOVA) which is used to compare the means of three or more groups. It explains that ANOVA avoids the problems of multiple t-tests by providing an omnibus test of differences between groups. The key steps of ANOVA are outlined, including partitioning variation between and within groups to calculate an F-ratio. A large F value indicates more difference between groups than expected by chance alone.
1. The document presents 11 claims and then states that all of them are false. It emphasizes the importance of the scientific method to verify answers rather than accepting claims without evidence.
2. People's beliefs and knowledge can come from many sources like rumors, parents, friends, and experiences that are not always accurate. The scientific method helps minimize errors by standardizing steps that can be replicated.
3. The last part of the document discusses identifying problems, forming testable hypotheses, designing studies, collecting and analyzing data, and reporting findings to help verify answers through the scientific process.
This document discusses research designs and their implications for causal interpretation. It covers experimental and non-experimental designs, between-group and within-group designs, and the importance of initial equivalence through random assignment and ongoing equivalence through control of confounding variables. While true experiments can potentially support causal claims, issues like failed randomization or unaccounted confounds mean their results may still not be causally interpretable. Non-experiments provide only associative evidence due to lack of experimental control.
BUS 308 Week 3 Lecture 1 Examining Differences - Continued.docxcurwenmichaela
BUS 308 Week 3 Lecture 1
Examining Differences - Continued
Expected Outcomes
After reading this lecture, the student should be familiar with:
1. Issues around multiple testing
2. The basics of the Analysis of Variance test
3. Determining significant differences between group means
4. The basics of the Chi Square Distribution.
Overview
Last week, we found out ways to examine differences between a measure taken on two
groups (two-sample test situation) as well as comparing that measure to a standard (a one-sample
test situation). We looked at the F test which let us test for variance equality. We also looked at
the t-test which focused on testing for mean equality. We noted that the t-test had three distinct
versions, one for groups that had equal variances, one for groups that had unequal variances, and
one for data that was paired (two measures on the same subject, such as salary and midpoint for
each employee). We also looked at how the 2-sample unequal t-test could be used to use Excel
to perform a one-sample mean test against a standard or constant value. This week we expand
our tool kit to let us compare multiple groups for similar mean values.
A second tool will let us look at how data values are distributed – if graphed, would they
look the same? Different shapes or patterns often means the data sets differ in significant ways
that can help explain results.
Multiple Groups
As interesting as comparing two groups is, often it is a bit limiting as to what it tells us.
One obvious issue that we are missing in the comparisons made last week was equal work. This
idea is still somewhat hard to get a clear handle on. Typically, as we look at this issue, questions
arise about things such as performance appraisal ratings, education distribution, seniority impact,
etc.
Some of these can be tested with the tools introduced last week. We can see, for
example, if the performance rating average is the same for each gender. What we couldn’t do, at
this point however, is see if performance ratings differ by grade, do the more senior workers
perform relatively better? Is there a difference between ratings for each gender by grade level?
The same questions can be asked about seniority impact. This week will give us tools to expand
how we look at the clues hidden within the data set about equal pay for equal work.
ANOVA
So, let’s start taking a look at these questions. The first tool for this week is the Analysis
of Variance – ANOVA for short. ANOVA is often confusing for students; it says it analyzes
variance (which it does) but the purpose of an ANOVA test is to determine if the means of
different groups are the same! Now, so far, we have considered means and variance to be two
distinct characteristics of data sets; characteristics that are not related, yet here we are saying that
looking at one will give us insight into the other.
The reason is due to the way the variance is an.
This document provides an overview of a one-way analysis of variance (ANOVA). It defines a one-way ANOVA as used to compare group means on a continuous dependent variable when there are two or more independent groups. Key steps outlined include calculating sums of squares between and within groups to partition total variability, computing the F ratio test statistic, and comparing this value to a critical value from the F distribution to determine if group means differ significantly. Factors that influence statistical significance, such as increasing between-group differences or decreasing within-group variability, are also discussed.
Levels of an independent variable (Practice)Ken Plummer
The document presents 5 practice problems involving determining the number of levels of the independent variable in different research studies. The problems include studies looking at: 1) student lunch requests and GPA with 2 levels, 2) religious beliefs and depression with 2 levels, 3) suicidal thoughts across grades with 4 levels, 4) college acceptance rates across gender with 2 levels, and 5) time management skills before and after a course with 2 levels.
Data.savQuestion.docxOver the same period, determine wheth.docxtheodorelove43763
This document provides instructions for conducting three chi-square tests of independence using SPSS on data about students' conflict resolution styles and suspensions from school. It describes entering the data, selecting the appropriate tests and variables, and interpreting the output. Students are asked to conduct the chi-square tests following the five steps of hypothesis testing, calculate effect sizes, and explain the results to someone unfamiliar with statistics.
Similar to What is a one-way repeated measures ANOVA? (20)
Diff rel gof-fit - jejit - practice (5)Ken Plummer
The document discusses the differences between questions of difference, relationship, and goodness of fit. It provides examples to illustrate each type of question. A question of difference compares two or more groups on some outcome, like comparing younger and older drivers' average driving speeds. A question of relationship examines whether a change in one variable causes a change in another, such as the relationship between age and flexibility. A question of goodness of fit assesses how well a claim matches reality, such as whether a salesman's claim of software effectiveness fits the results of user testing.
This document provides examples of questions that ask for the lowest and highest number in a set of data. The questions ask for the difference between the state with the lowest and highest church attendance, the students with the highest and lowest test scores, and the slowest and fastest versions of a vehicle model.
Inferential vs descriptive tutorial of when to use - Copyright UpdatedKen Plummer
The document discusses the differences between descriptive and inferential statistics. Descriptive statistics are used to describe characteristics of a whole population, while inferential statistics are used when the whole population cannot be measured and conclusions are drawn from a sample to generalize to the larger population. Examples are provided to illustrate when each type of statistic would be used. Key differences include descriptive statistics examining entire populations while inferential statistics examine samples that aim to infer conclusions about populations.
Diff rel ind-fit practice - Copyright UpdatedKen Plummer
The document provides explanations and examples for different types of statistical questions:
- Difference questions compare two or more groups on an outcome.
- Relationship questions examine if a change in one variable is associated with a change in another variable.
- Independence questions determine if two variables with multiple levels are independent of each other.
- Goodness of fit questions assess how well a claim matches reality.
Examples are given for each type of question to illustrate key concepts like comparing groups, examining associations between variables, assessing independence, and evaluating how a claim fits observed data.
Normal or skewed distributions (inferential) - Copyright updatedKen Plummer
- The document discusses determining whether distributions are normal or skewed
- A distribution is considered skewed if the skewness value divided by the standard error of skewness is less than -2 or greater than 2
- For the old car data set in the example, the skewness value of -4.26 divided by the standard error is less than -2, so this distribution is negatively skewed
- The new car data set skewness value of -1.69 divided by the standard error is between -2 and 2, so this distribution is normal
Normal or skewed distributions (descriptive both2) - Copyright updatedKen Plummer
The document discusses normal and skewed distributions and how to identify them. It provides examples of measuring forearm circumference of golf players and IQs of cats and dogs. The forearm circumference data is normally distributed while the dog IQ data is left skewed based on the skewness statistics provided. Therefore, at least one of the distributions (dog IQs) is skewed.
Nature of the data practice - Copyright updatedKen Plummer
The document discusses different types of data:
- Scaled data provides exact amounts like 12.5 feet or 140 miles per hour.
- Ordinal or ranked data provides comparative amounts like 1st, 2nd, 3rd place.
- Nominal data names or categorizes values like Republican or Democrat.
- Nominal proportional data are simply percentages like Republican 45% or Democrat 55%.
Nature of the data (spread) - Copyright updatedKen Plummer
The document discusses scaled and ordinal data. Scaled data can be measured in exact amounts like distances and speeds. Ordinal data provides comparative amounts by ranking items, like the top 3 states in terms of well-being. Examples ask the reader to identify if data is scaled or ordinal, like driving speeds which are scaled, or baby weight percentiles which are ordinal as they compare weights.
The document is a series of questions and examples that explain what it means for a question to ask about the "most frequent response". It provides examples of questions asking about the highest/most number of something based on data in tables or lists. It then asks a series of questions to determine if they are asking about the most frequent/common response based on the data given.
Nature of the data (descriptive) - Copyright updatedKen Plummer
The document discusses two types of data: scaled data and ordinal data. Scaled data can be measured in exact amounts with equal intervals between values. Ordinal or ranked data provides comparative amounts but not necessarily equal intervals. Several examples are provided to illustrate the difference, including driving speed, states ranked by well-being, and elephant weights. Practice questions are also included for the reader to determine if data examples provided are scaled or ordinal.
The document discusses whether variables are dichotomous or scaled when calculating correlations. It provides examples of correlations between ACT scores and whether students attended private or public school. One example has ACT scores as a scaled variable and school type as dichotomous. Another has lower and higher ACT scores as dichotomous and school type as dichotomous. It emphasizes determining if variables are both dichotomous, or if one is dichotomous and one is scaled.
The document discusses the correlation between ACT scores and a measure of school belongingness. It determines that one of the variables, which has a sample size less than 30, is skewed and has many ties. As a result, a non-parametric test should be used to analyze the relationship between the two variables.
The document discusses using parametric versus non-parametric tests based on sample size for skewed distributions. For skewed distributions with a sample size less than 30, a non-parametric test is recommended. For skewed distributions with a sample size greater than or equal to 30, a parametric test is recommended. It provides examples analyzing the correlation between ACT scores and sense of school belongingness using both approaches.
The document discusses whether there are many ties or few/no ties within the variables of the relationship question "What is the correlation between ACT rankings (ordinal) and sense of school belongingness (scaled 1-10)?". It determines that ACT rankings, being ordinal, have many ties, while sense of school belongingness, being on a scale of 1-10, may have many or few ties depending on how scores are distributed.
The document discusses identifying whether variables in statistical analyses are ordinal or nominal. It provides examples of relationships between variables such as ACT rankings and sense of school belongingness, daily social media use and sense of well-being, and private/public school enrollment and sense of well-being. It asks the reader to identify if variables in examples like running speed and shoe/foot size or LSAT scores and test anxiety are ordinal or nominal.
The document discusses covariates and their impact on relationships between variables. It defines a covariate as a variable that is controlled for or eliminated from a study. It explains that if a covariate is related to one of the variables in the relationship being examined, it can impact the strength of that relationship. Examples are provided to demonstrate when a question involves a covariate or not.
This document discusses the nature of variables in relationship questions. It can be determined that the variables are either both scaled, at least one is ordinal, or at least one is nominal. Examples of different relationship questions are provided that fall into each of these categories. The document also provides practice questions for the user to determine which category the variables fall into.
The document discusses the number of variables involved in research questions. It explains that many relationship questions deal with two variables, such as gender predicting driving speed. However, some questions deal with three or more variables, for example gender and age predicting driving speed. The document asks the reader to identify whether example research questions involve two or three or more variables.
The document discusses independent and dependent variables in research questions. It provides examples to illustrate that an independent variable has at least two levels and may have more, such as religious affiliation having two levels (Western religion and Eastern religion) or company type having three levels (Company X, Company Y, Company Z). It then provides a practice example about employee satisfaction rates among morning, afternoon, and evening shifts, identifying shift status as the independent variable with three levels.
The document discusses independent variables and how they relate to research questions. It provides examples of questions with one independent variable, two independent variables, and zero independent variables. An independent variable influences or impacts a dependent variable. Questions are presented about employee satisfaction rates, agent commissions, training proficiency, and cyberbullying incidents to illustrate different numbers of independent variables.
Chapter wise All Notes of First year Basic Civil Engineering.pptxDenish Jangid
Chapter wise All Notes of First year Basic Civil Engineering
Syllabus
Chapter-1
Introduction to objective, scope and outcome the subject
Chapter 2
Introduction: Scope and Specialization of Civil Engineering, Role of civil Engineer in Society, Impact of infrastructural development on economy of country.
Chapter 3
Surveying: Object Principles & Types of Surveying; Site Plans, Plans & Maps; Scales & Unit of different Measurements.
Linear Measurements: Instruments used. Linear Measurement by Tape, Ranging out Survey Lines and overcoming Obstructions; Measurements on sloping ground; Tape corrections, conventional symbols. Angular Measurements: Instruments used; Introduction to Compass Surveying, Bearings and Longitude & Latitude of a Line, Introduction to total station.
Levelling: Instrument used Object of levelling, Methods of levelling in brief, and Contour maps.
Chapter 4
Buildings: Selection of site for Buildings, Layout of Building Plan, Types of buildings, Plinth area, carpet area, floor space index, Introduction to building byelaws, concept of sun light & ventilation. Components of Buildings & their functions, Basic concept of R.C.C., Introduction to types of foundation
Chapter 5
Transportation: Introduction to Transportation Engineering; Traffic and Road Safety: Types and Characteristics of Various Modes of Transportation; Various Road Traffic Signs, Causes of Accidents and Road Safety Measures.
Chapter 6
Environmental Engineering: Environmental Pollution, Environmental Acts and Regulations, Functional Concepts of Ecology, Basics of Species, Biodiversity, Ecosystem, Hydrological Cycle; Chemical Cycles: Carbon, Nitrogen & Phosphorus; Energy Flow in Ecosystems.
Water Pollution: Water Quality standards, Introduction to Treatment & Disposal of Waste Water. Reuse and Saving of Water, Rain Water Harvesting. Solid Waste Management: Classification of Solid Waste, Collection, Transportation and Disposal of Solid. Recycling of Solid Waste: Energy Recovery, Sanitary Landfill, On-Site Sanitation. Air & Noise Pollution: Primary and Secondary air pollutants, Harmful effects of Air Pollution, Control of Air Pollution. . Noise Pollution Harmful Effects of noise pollution, control of noise pollution, Global warming & Climate Change, Ozone depletion, Greenhouse effect
Text Books:
1. Palancharmy, Basic Civil Engineering, McGraw Hill publishers.
2. Satheesh Gopi, Basic Civil Engineering, Pearson Publishers.
3. Ketki Rangwala Dalal, Essentials of Civil Engineering, Charotar Publishing House.
4. BCP, Surveying volume 1
Temple of Asclepius in Thrace. Excavation resultsKrassimira Luka
The temple and the sanctuary around were dedicated to Asklepios Zmidrenus. This name has been known since 1875 when an inscription dedicated to him was discovered in Rome. The inscription is dated in 227 AD and was left by soldiers originating from the city of Philippopolis (modern Plovdiv).
Beyond Degrees - Empowering the Workforce in the Context of Skills-First.pptxEduSkills OECD
Iván Bornacelly, Policy Analyst at the OECD Centre for Skills, OECD, presents at the webinar 'Tackling job market gaps with a skills-first approach' on 12 June 2024
A Visual Guide to 1 Samuel | A Tale of Two HeartsSteve Thomason
These slides walk through the story of 1 Samuel. Samuel is the last judge of Israel. The people reject God and want a king. Saul is anointed as the first king, but he is not a good king. David, the shepherd boy is anointed and Saul is envious of him. David shows honor while Saul continues to self destruct.
This presentation was provided by Rebecca Benner, Ph.D., of the American Society of Anesthesiologists, for the second session of NISO's 2024 Training Series "DEIA in the Scholarly Landscape." Session Two: 'Expanding Pathways to Publishing Careers,' was held June 13, 2024.
3. How did you get here?
So, you have decided to use a Repeated
Measures ANOVA.
4. How did you get here?
So, you have decided to use a Repeated
Measures ANOVA.
Let’s consider the decisions you made to get
here.
5. First of all, you must have noticed the problem
to be solved deals with generalizing from a
smaller sample to a larger population.
6. First of all, you must have noticed the problem
to be solved deals with generalizing from a
smaller sample to a larger population.
7. First of all, you must have noticed the problem
to be solved deals with generalizing from a
smaller sample to a larger population.
Sample of 30
8. First of all, you must have noticed the problem
to be solved deals with generalizing from a
smaller sample to a larger population.
Sample of 30
9. First of all, you must have noticed the problem
to be solved deals with generalizing from a
smaller sample to a larger population.
Large Population of 30,000
Sample of 30
10. First of all, you must have noticed the problem
to be solved deals with generalizing from a
smaller sample to a larger population.
Large Population of 30,000
Sample of 30
Therefore, you would determine that the
problem deals with inferential not descriptive
statistics.
11. Therefore, you would determine that the
problem deals with inferential not descriptive
statistics.
12. Therefore, you would determine that the
problem deals with inferential not descriptive
statistics.
Double check your
problem to see if
that is the case
13. Therefore, you would determine that the
problem deals with inferential not descriptive
statistics.
Inferential Descriptive
Double check your
problem to see if
that is the case
14. You would have also noticed that the problem
dealt with questions of difference not
Relationships, Independence nor Goodness of
Fit. Inferential Descriptive
15. You would have also noticed that the problem
dealt with questions of difference not
Relationships, Independence nor Goodness of
Fit.
Double check your
problem to see if
that is the case
Inferential Descriptive
Difference
16. You would have also noticed that the problem
dealt with questions of difference not
Relationships, Independence nor Goodness of
Fit.
Double check your
problem to see if
that is the case
Inferential Descriptive
Difference Relationship
17. You would have also noticed that the problem
dealt with questions of difference not
Relationships, Independence nor Goodness of
Fit.
Double check your
problem to see if
that is the case
Inferential Descriptive
Difference Relationship Difference
18. You would have also noticed that the problem
dealt with questions of difference not
Relationships, Independence nor Goodness of
Fit.
Double check your
problem to see if
that is the case
Inferential Descriptive
Difference Relationship Difference Goodness of Fit
19. After checking the data, you noticed that the
data was ratio/interval rather than extreme
ordinal (1st, 2nd, 3rd place) or nominal (male,
female)
Double check your
problem to see if
that is the case
Inferential Descriptive
Difference Relationship Difference Goodness of Fit
20. After checking the data, you noticed that the
data was ratio/interval rather than extreme
ordinal (1st, 2nd, 3rd place) or nominal (male,
female)
Double check your
problem to see if
that is the case
Inferential Descriptive
Difference Goodness of Fit
Difference Relationship
Ratio/Interval
21. After checking the data, you noticed that the
data was ratio/interval rather than extreme
ordinal (1st, 2nd, 3rd place) or nominal (male,
female)
Double check your
problem to see if
that is the case
Inferential Descriptive
Difference Goodness of Fit
Difference Relationship
Ratio/Interval Ordinal
22. After checking the data, you noticed that the
data was ratio/interval rather than extreme
ordinal (1st, 2nd, 3rd place) or nominal (male,
female)
Double check your
problem to see if
that is the case
Inferential Descriptive
Difference Goodness of Fit
Difference Relationship
Ratio/Interval Ordinal Nominal
27. The distribution was more or less normal rather
than skewed or kurtotic.
Double check your
problem to see if
that is the case
Inferential Descriptive
Difference Goodness of Fit
Skewed
Difference Relationship
Ratio/Interval Ordinal Nominal
28. The distribution was more or less normal rather
than skewed or kurtotic.
Double check your
problem to see if
that is the case
Inferential Descriptive
Difference Goodness of Fit
Difference Relationship
Ratio/Interval Ordinal Nominal
Skewed Kurtotic
29. The distribution was more or less normal rather
than skewed or kurtotic.
Double check your
problem to see if
that is the case
Inferential Descriptive
Difference Goodness of Fit
Difference Relationship
Ratio/Interval Ordinal Nominal
Skewed Kurtotic Normal
31. Only one Dependent Variable (DV) rather than
two or more exist.
DV #1
Chemistry
Test Scores
32. Only one Dependent Variable (DV) rather than
two or more exist.
DV #1 DV #2
Chemistry
Test Scores
Class
Attendance
33. Only one Dependent Variable (DV) rather than
two or more exist.
DV #1 DV #2 DV #3
Chemistry
Test Scores
Class
Attendance
Homework
Completed
34. Only one Dependent Variable (DV) rather than
two or more exist.
Inferential Descriptive
Difference Goodness of Fit
Double check your
problem to see if
that is the case
Difference Relationship
Ratio/Interval Ordinal Nominal
Skewed Kurtotic Normal
35. Only one Dependent Variable (DV) rather than
two or more exist.
Descriptive
Difference Goodness of Fit
Difference Relationship
Skewed Kurtotic Normal
1 DV
Double check your
problem to see if
that is the case
Inferential
Ratio/Interval Ordinal Nominal
36. Only one Dependent Variable (DV) rather than
two or more exist.
Inferential Descriptive
Difference Relationship Difference Goodness of Fit
Ratio/Interval Ordinal Nominal
Skewed Kurtotic Normal
1 DV 2+ DV
Double check your
problem to see if
that is the case
38. Only one Independent Variable (DV) rather than
two or more exist.
IV #1
Use of Innovative
eBook
39. Only one Independent Variable (DV) rather than
two or more exist.
IV #1 IV #2
Use of Innovative
eBook
Doing Homework
to Classical Music
40. Only one Independent Variable (DV) rather than
two or more exist.
IV #1 IV #2 IV #3
Use of Innovative
eBook
Doing Homework
to Classical Music
Gender
41. Only one Independent Variable (DV) rather than
two or more exist.
IV #1 IV #2 IV #3
Use of Innovative
eBook
Doing Homework
to Classical Music
Gender
43. Only one Independent Variable (DV) rather than
two or more exist. Descriptive
Difference Goodness of Fit
Inferential
Difference Relationship
Ratio/Interval Ordinal Nominal
Skewed Kurtotic Normal
1 DV 2+ DV
44. Only one Independent Variable (DV) rather than
two or more exist. Inferential Descriptive
Difference Goodness of Fit
Difference Relationship
Skewed Kurtotic Normal
1 DV 2+ DV
1 IV
Inferential
Ratio/Interval Ordinal Nominal
45. Only one Independent Variable (DV) rather than
two or more exist. Descriptive
Difference Relationship Difference
Difference Goodness of Fit
Nominal
Skewed Kurtotic Normal
1 DV 2+ DV
1 IV 2+ IV
Inferential
Ratio/Interval Ordinal Nominal
46. Only one Independent Variable (DV) rather than
two or more exist. Descriptive
Difference Relationship Difference
Difference Goodness of Fit
Skewed Kurtotic Normal
1 DV 2+ DV
1 IV 2+ IV
Double check your
problem to see if
that is the case
Inferential
Ratio/Interval Ordinal Nominal
47. There are three levels of the Independent
Variable (IV) rather than just two levels. Note –
even though repeated measures ANOVA can
analyze just two levels, this is generally analyzed
using a paired sample t-test.
48. There are three levels of the Independent
Variable (DV) rather than just two levels. Note –
even though repeated measures ANOVA can
analyze just two levels, this is generally analyzed
using a paired sample t-test.
Level 1
Before using the
innovative ebook
49. There are three levels of the Independent
Variable (DV) rather than just two levels. Note –
even though repeated measures ANOVA can
analyze just two levels, this is generally analyzed
using a paired sample t-test.
Level 1 Level 2
Before using the
innovative ebook
Using the
innovative ebook
for 2 months
50. There are three levels of the Independent
Variable (DV) rather than just two levels. Note –
even though repeated measures ANOVA can
analyze just two levels, this is generally analyzed
using a paired sample t-test.
Level 1 Level 2 Level 3
Before using the
innovative ebook
Using the
innovative ebook
for 2 months
Using the
innovative ebook
for 4 months
51. Descriptive
Goodness of Fit
Difference Relationship
Skewed Kurtotic Normal
1 DV 2+ DVs
2+ IVs
Inferential
Ratio/Interval Ordinal Nominal
1 IV
2 levels 3+ levels
Difference
52. The samples are repeated rather than
independent. Notice that the same class (Chem
100 section 003) is repeatedly tested.
53. The samples are repeated rather than
independent. Notice that the same class (Chem
100 section 003) is repeatedly tested.
Chem 100
Section 003
January
Chem 100
Section 003
March
Chem 100
Section 003
May
Before using
the innovative
ebook
Using the
innovative ebook
for 2 months
Using the
innovative ebook
for 4 months
54. Descriptive
Goodness of Fit
Difference Relationship
Skewed Kurtotic Normal
1 DV 2+ DVs
2+ IVs
Inferential
Ratio/Interval Ordinal Nominal
1 IV
2 levels 3+ levels
Difference
Independent Repeated
55. If this was the appropriate path for your
problem then you have correctly selected
Repeated-measures ANOVA to solve the
problem you have been presented.
57. Repeated Measures ANOVA –
Another use of analysis of variance is to test
whether a single group of people change over
time.
58. Repeated Measures ANOVA –
Another use of analysis of variance is to test
whether a single group of people change over
time.
59. In this case, the distributions that are compared
to each other are not from different groups
60. In this case, the distributions that are compared
to each other are not from different groups
versus
Group 1 Group 2
61. In this case, the distributions that are compared
to each other are not from different groups
versus
Group 1 Group 2
62. In this case, the distributions that are compared
to each other are not from different groups
versus
Group 1 Group 2
But from different times.
63. In this case, the distributions that are compared
to each other are not from different groups
versus
Group 1 Group 2
But from different times.
Group 1 Group 1:
Two Months Later
versus
64. For example, an instructor might administer the
same test three times throughout the semester
to ascertain whether students are improving in
their skills.
65. For example, an instructor might administer the
same test three times throughout the semester
to ascertain whether students are improving in
their skills.
January February
April
Exam 1
Exam 2
Exam 3
66. For example, an instructor might administer the
same test three times throughout the semester
to ascertain whether students are improving in
their skills.
Exam 1
Exam 2
January February
Exam 3
April
The overall F-ratio will reveal whether there are
differences somewhere among three time
periods.
67. For example, an instructor might administer the
same test three times throughout the semester
to ascertain whether students are improving in
their skills.
Exam 1
Exam 2
January February
Exam 3
April
The overall F-ratio will reveal whether there are
differences somewhere among three time
periods.
68. For example, an instructor might administer the
same test three times throughout the semester
to ascertain whether students are improving in
their skills.
Exam 1
Exam 2
Average
Score
January February
Exam 3
April
Average
Score
Average
Score
The overall F-ratio will reveal whether there are
differences somewhere among three time
periods.
69. For example, an instructor might administer the
same test three times throughout the semester
to ascertain whether students are improving in
their skills.
Exam 1
Exam 2
Average
Score
January February
Exam 3
April
Average
Score
Average
Score
The overall F-ratio will reveal whether there are
differences somewhere among three time
periods.
70. For example, an instructor might administer the
same test three times throughout the semester
to ascertain whether students are improving in
their skills.
Exam 1
Exam 2
Average
Score
January February
Exam 3
April
Average
Score
Average
Score
There is a
difference but
we don’t
know where
The overall F-ratio will reveal whether there are
differences somewhere among three time
periods.
71. Post hoc tests will reveal exactly where the
differences occurred.
72. Post hoc tests will reveal exactly where the
differences occurred.
January February
April
Exam 1
Exam 2
Exam 3
Average
Score 35
Average
Score 38
Average
Score 40
73. Post hoc tests will reveal exactly where the
differences occurred.
January February
April
Exam 1
Exam 2
Exam 3
Average
Score 35
Average
Score 38
Average
Score 40
There is a
statistically
significant
difference only
between Exam 1
and Exam 3
74. In contrast, with the One-way analysis of
Variance (ANOVA) we were attempting to
determine if there was a statistical difference
between 2 or more (generally 3 or more)
groups.
75. In contrast, with the One-way analysis of
Variance (ANOVA) we were attempting to
determine if there was a statistical difference
between 2 or more (generally 3 or more)
groups.
In our One-way ANOVA example in another
presentation we attempted to determine if
there was any statistically significant difference
in the amount of Pizza Slices consumed by three
different player types (football, basketball, and
soccer).
77. The data would be set up thus:
Football
Players
Pizza
Slices
Consumed
Basketball
Players
Pizza Slices
Consumed
Soccer
Players
Pizza Slices
Consumed
Ben 5 Cam 6 Dan 5
Bob 7 Colby 4 Denzel 8
Bud 8 Conner 8 Dilbert 8
Bubba 9 Custer 4 Don 1
Burt 10 Cyan 2 Dylan 2
78. The data would be set up thus:
Football
Players
Pizza
Slices
Consumed
Basketball
Players
Pizza Slices
Consumed
Soccer
Players
Pizza Slices
Consumed
Ben 5 Cam 6 Dan 5
Bob 7 Colby 4 Denzel 8
Bud 8 Conner 8 Dilbert 8
Bubba 9 Custer 4 Don 1
Burt 10 Cyan 2 Dylan 2
Notice how the individuals in these groups are
different (hence different names)
79. The data would be set up thus:
Football
Players
Pizza
Slices
Consumed
Basketball
Players
Pizza Slices
Consumed
Soccer
Players
Pizza Slices
Consumed
Ben 5 Cam 6 Dan 5
Bob 7 Colby 4 Denzel 8
Bud 8 Conner 8 Dilbert 8
Bubba 9 Custer 4 Don 1
Burt 10 Cyan 2 Dylan 2
Notice how the individuals in these groups are
different (hence different names)
80. The data would be set up thus:
Football
Players
Pizza
Slices
Consumed
Basketball
Players
Pizza Slices
Consumed
Soccer
Players
Pizza Slices
Consumed
Ben 5 Ben 6 Ben 5
Bob 7 Bob 4 Bob 8
Bud 8 Bud 8 Bud 8
Bubba 9 Bubba 4 Bubba 1
Burt 10 Burt 2 Burt 2
Notice how the individuals in these groups are
different (hence different names)
A Repeated Measures ANOVA is different than a
One-Way ANOVA in one simply way: Only one
group of person or observations is being
measured, but they are measured more than
one time.
81. The data would be set up thus:
Football
Players
Pizza
Slices
Consumed
Basketball
Players
Pizza Slices
Consumed
Soccer
Players
Pizza Slices
Consumed
Ben 5 Ben 6 Ben 5
Bob 7 Bob 4 Bob 8
Bud 8 Bud 8 Bud 8
Bubba 9 Bubba 4 Bubba 1
Burt 10 Burt 2 Burt 2
Notice how the individuals in these groups are
different (hence different names)
A Repeated Measures ANOVA is different than a
One-Way ANOVA in one simply way: Only one
group of persons or observations is being
measured, but they are measured more than
one time.
82. Notice the different times football player pizza
consumption is being measured.
Football
Players
Pizza
Slices
Consumed
Pizza Slices
Consumed
Pizza Slices
Consumed
Ben 5 Ben 6 Ben 5
Bob 7 Bob 4 Bob 8
Bud 8 Bud 8 Bud 8
Bubba 9 Bubba 4 Bubba 1
Burt 10 Burt 2 Burt 2
83. Notice the different times football player pizza
consumption is being measured.
Football
Players
Pizza
Slices
Consumed
Before the
Season
Pizza Slices
Consumed
During the
Season
Pizza Slices
Consumed
After the
Season
Ben 5 Ben 6 Ben 5
Bob 7 Bob 4 Bob 8
Bud 8 Bud 8 Bud 8
Bubba 9 Bubba 4 Bubba 1
Burt 10 Burt 2 Burt 2
84. Since only one group is being measured 3 times,
each time is dependent on the previous time.
By dependent we mean there is a relationship.
85. Since only one group is being measured 3 times,
each time is dependent on the previous time.
By dependent we mean there is a relationship.
Pizza Slices Consumed
Football Players Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
86. Since only one group is being measured 3 times,
each time is dependent on the previous time.
By dependent we mean there is a relationship.
Pizza Slices Consumed
Football Players Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
The relationship between the scores is that we
are comparing the same person across multiple
observations.
87. So, Ben’s before-season and during-season and
after-season scores have one important thing in
common:
88. So, Ben’s before-season and during-season and
after-season scores have one important thing in
common:
Pizza Slices Consumed
Football Players Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
89. So, Ben’s before-season and during-season and
after-season scores have one important thing in
common: THESE SCORES ALL BELONG TO BEN.
Pizza Slices Consumed
Football Players Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
90. So, Ben’s before-season and during-season and
after-season scores have one important thing in
common: THESE SCORES ALL BELONG TO BEN.
Pizza Slices Consumed
Football Players Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
They are subject to all the factors that are
special to Ben when consuming pizza, including
how much he likes or dislikes, the toppings that
are available, the eating atmosphere, etc.
91. What we want to find out is – how much the
BEFORE, DURING, and AFTER season pizza
consuming sessions differ.
92. What we want to find out is – how much the
BEFORE, DURING, and AFTER season pizza
consuming sessions differ.
But we have to find a way to eliminate the
variability that is caused by individual
differences that linger across all three eating
sessions. Once again we are not interested in
the things that make Ben, Ben while eating pizza
(like he’s a picky eater). We are interested in the
effect of where we are in the season (BEFORE,
DURING, and AFTER on Pizza consumption.)
93. What we want to find out is – how much the
BEFORE, DURING, and AFTER season pizza
consuming sessions differ.
But we have to find a way to eliminate the
variability that is caused by individual
differences that linger across all three eating
sessions. Once again we are not interested in
the things that make Ben, Ben while eating pizza
(like he’s a picky eater). We are interested in the
effect of where we are in the season (BEFORE,
DURING, and AFTER on Pizza consumption.)
94. What we want to find out is – how much the
BEFORE, DURING, and AFTER season pizza
consuming sessions differ.
But we have to find a way to eliminate the
variability that is caused by individual
differences that linger across all three eating
sessions. Once again we are not interested in
the things that make Ben, Ben while eating pizza
(like he’s a picky eater). We are interested in the
effect of where we are in the season (BEFORE,
DURING, and AFTER on Pizza consumption.)
95. That way we can focus just on the differences
that are related to WHEN the pizza eating
occurred.
96. That way we can focus just on the differences
that are related to WHEN the pizza eating
occurred.
After running a repeated-measures ANOVA, this
is the output that we will get:
97. That way we can focus just on the differences
that are related to WHEN the pizza eating
occurred.
After running a repeated-measures ANOVA, this
is the output that we will get:
Tests of Within-Subjects Effects
Measure: Pizza slices
Source
Type III
Sum of
Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
98. This output will help us determine if we reject
the null hypothesis:
99. This output will help us determine if we reject
the null hypothesis:
There is no significant difference in the amount
of pizza consumed by football players before,
during, and/or after the season.
100. This output will help us determine if we reject
the null hypothesis:
There is no significant difference in the amount
of pizza consumed by football players before,
during, and/or after the season.
Or accept the alternative hypothesis:
101. This output will help us determine if we reject
the null hypothesis:
There is no significant difference in the amount
of pizza consumed by football players before,
during, and/or after the season.
Or accept the alternative hypothesis:
There is a significant difference in the amount of
pizza consumed by football players before,
during, and/or after the season.
103. To do so, let’s focus on the value .008
Tests of Within-Subjects Effects
Measure: Pizza slices consumed
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
104. To do so, let’s focus on the value .008
Tests of Within-Subjects Effects
Measure:
Pizza slices
consumed
Source
Type III
Sum of
Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
105. To do so, let’s focus on the value .008
Tests of Within-Subjects Effects
Measure:
Pizza slices
consumed
Source
Type III
Sum of
Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
This means that if we were to reject the null
hypothesis, the probability that we would be
wrong is 8 times out of 1000. As you remember,
if that were to happen, it would be called a Type
1 error.
106. To do so, let’s focus on the value .008
Tests of Within-Subjects Effects
Measure:
Pizza slices
consumed
Source
Type III
Sum of
Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
This means that if we were to reject the null
hypothesis, the probability that we would be
wrong is 8 times out of 1000. As you remember,
if that were to happen, it would be called a Type
1 error.
107. But it is so unlikely, that we would be willing to
take that risk and hence reject the null
hypothesis.
108. But it is so unlikely, that we would be willing to
take that risk and hence we reject the null
hypothesis.
There IS NO statistically significant difference
between the number of slices of pizza
consumed by football players before, during, or
after the football season.
109. But it is so unlikely, that we would be willing to
take that risk and hence we reject the null
hypothesis.
There IS NO statistically significant difference
between the number of slices of pizza
consumed by football players before, during, or
after the football season.
111. And accept the alternative hypothesis:
There IS A statistically significant difference
between the number of slices of pizza
consumed by football players before, during, or
after the football season.
112. And accept the alternative hypothesis:
There IS A statistically significant difference
between the number of slices of pizza
consumed by football players before, during, or
after the football season.
113. Now we do not know which of the three are
significantly different from one another or if all
three are different. We just know that a
difference exists.
114. Now we do not know which of the three are
significantly different from one another or if all
three are different. We just know that a
difference exists.
Pizza Slices Consumed
Football Players Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
115. Now we do not know which of the three are
significantly different from one another or if all
three are different. We just know that a
difference exists.
Pizza Slices Consumed
Football Players Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
116. Now we do not know which of the three are
significantly different from one another or if all
three are different. We just know that a
difference exists.
Pizza Slices Consumed
Football Players Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
Later, we can run what is called a “Post-hoc” test
to determine where the difference lies.
117. From this point on – we will delve into the
actual calculations and formulas that produce a
Repeated-measures ANOVA. If such detail is of
interest or a necessity to know, please continue.
118. How was a significance value of .008 calculated?
119. How was a significance value of .008 calculated?
Let’s begin with the calculation of the various
sources of Sums of Squares
120. How was a significance value of .008 calculated?
Let’s begin with the calculation of the various
sources of Sums of Squares
Tests of Within-Subjects Effects
Measure:
Pizza slices
consumed
Source
Type III
Sum of
Squares df
Mean
Square F Sig.
Between
Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
121. We do this so that we can explain what is
causing the scores to vary or deviate.
122. We do this so that we can explain what is
causing the scores to vary or deviate.
• Is it error?
123. We do this so that we can explain what is
causing the scores to vary or deviate.
• Is it error?
• Is it differences between times (before,
during, and after)?
124. We do this so that we can explain what is
causing the scores to vary or deviate.
• Is it error?
• Is it differences between times (before,
during, and after)?
Remember, the full name for sum of squares is
the sum of squared deviations about the mean.
This will help us determine the amount of
variation from each of the possible sources.
126. Let’s begin by calculating the total sums of
squares.
푆푆푡표푡푎푙 = Σ(푋푖푗 − 푋 )2
127. Let’s begin by calculating the total sums of
squares.
푆푆푡표푡푎푙 = Σ(푋푖푗 − 푋 )2
128. Let’s begin by calculating the total sums of
squares.
푆푆푡표푡푎푙 = Σ(푋푖푗 − 푋 )2
This means one pizza
eating observation for
person “I” (e.g., Ben) on
time “j” (e.g., before)
130. For example:
Pizza Slices Consumed
Football Players Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
131. For example:
Pizza Slices Consumed
Football Players Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
132. For example:
Pizza Slices Consumed
Football Players Before the
OR
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
133. For example:
Pizza Slices Consumed
Football Players Before the
OR
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
Pizza Slices Consumed
Football Players Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
134. For example:
Pizza Slices Consumed
Football Players Before the
OR
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
Pizza Slices Consumed
Football Players Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
135. For example:
Pizza Slices Consumed
Football Players Before the
OR
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
136. For example:
Pizza Slices Consumed
Football Players Before the
OR
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
Pizza Slices Consumed
Football Players Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
137. For example:
Pizza Slices Consumed
Football Players Before the
Season
During the
Season
ETC
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
139. 푆푆푡표푡푎푙 = Σ(푋푖푗 − 푿)2
This means the
average of all of the
observations
140. 푆푆푡표푡푎푙 = Σ(푋푖푗 − 푿)2
This means the
average of all of the
observations
This means one pizza
eating observation for
person “I” (e.g., Ben) on
time “j” (e.g., before)
141. 푆푆푡표푡푎푙 = Σ(푋푖푗 − 푿)2
This means the
average of all of the
observations
Pizza Slices Consumed
This means one pizza
eating observation for
person “I” (e.g., Ben) on
time “j” (e.g., before)
Football Players Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
142. 푆푆푡표푡푎푙 = Σ(푋푖푗 − 푿)2
This means the
average of all of the
observations
Pizza Slices Consumed
Football Players Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
Average of All
Observations
This means one pizza
eating observation for
person “I” (e.g., Ben) on
time “j” (e.g., before)
144. 푆푆푡표푡푎푙 = Σ(푋푖푗 − 푿
)2
This means
sum or add
everything up
This means
the average of
all of the
observations
145. 푆푆푡표푡푎푙 = Σ(푋푖푗 − 푿)2
This means
sum or add
everything up
This means
the average of
all of the
observations
This means one pizza
eating observation for
person “I” (e.g., Ben) on
time “j” (e.g., before)
147. Let’s calculate total sums of squares with this
data set:
Pizza Slices Consumed
Football Players Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
149. To do so we will rearrange the data like so:
Football
Players
Ben
Bob
Bud
Bubba
Burt
Ben
Bob
Bud
Bubba
Burt
Ben
Bob
Bud
Bubba
Burt
150. To do so we will rearrange the data like so:
Football
Players
Ben
Bob
Bud
Bubba
Burt
Ben
Bob
Bud
Bubba
Burt
Ben
Bob
Bud
Bubba
Burt
Season
Before
Before
Before
Before
Before
During
During
During
During
During
After
After
After
After
After
151. To do so we will rearrange the data like so:
Football
Players
Ben
Bob
Bud
Bubba
Burt
Ben
Bob
Bud
Bubba
Burt
Ben
Bob
Bud
Bubba
Burt
Season
Before
Before
Before
Before
Before
During
During
During
During
During
After
After
After
After
After
Slices of
Pizza
5
7
8
9
10
4
5
7
8
7
4
5
6
4
6
152. To do so we will rearrange the data like so:
We will subtract each of these
values from the grand mean,
square the result and sum them
all up.
Football
Players
Ben
Bob
Bud
Bubba
Burt
Ben
Bob
Bud
Bubba
Burt
Ben
Bob
Bud
Bubba
Burt
Season
Before
Before
Before
Before
Before
During
During
During
During
During
After
After
After
After
After
Slices of
Pizza
5
7
8
9
10
4
5
7
8
7
4
5
6
4
6
153. To do so we will rearrange the data like so:
We will subtract each of these
values from the grand mean,
square the result and sum them
all up.
Football
Players
Ben
Bob
Bud
Bubba
Burt
Ben
Bob
Bud
Bubba
Burt
Ben
Bob
Bud
Bubba
Burt
Season
Before
Before
Before
Before
Before
During
During
During
During
During
After
After
After
After
After
Slices of
Pizza
5
7
8
9
10
4
5
7
8
7
4
5
6
4
6 푆푆푡표푡푎푙 = Σ(푋푖푗 − 푋 )2
154. To do so we will rearrange the data like so:
We will subtract each of these
values from the grand mean,
square the result and sum them
all up.
Football
Players
Ben
Bob
Bud
Bubba
Burt
Ben
Bob
Bud
Bubba
Burt
Ben
Bob
Bud
Bubba
Burt
Season
Before
Before
Before
Before
Before
During
During
During
During
During
After
After
After
After
After
Slices of
Pizza
5
7
8
9
10
4
5
7
Each
8
7
observation
4
5
6
4
푆푆2
6 푡표푡푎푙 = Σ(푋푖푗 − 푋 )
155. To do so we will rearrange the data like so:
We will subtract each of these
values from the grand mean,
square the result and sum them
all up.
Here is how we compute the
Grand Mean =
Football
Players
Ben
Bob
Bud
Bubba
Burt
Ben
Bob
Bud
Bubba
Burt
Ben
Bob
Bud
Bubba
Burt
Season
Before
Before
Before
Before
Before
During
During
During
During
During
After
After
After
After
After
Slices of
Pizza
5
7
8
9
10
4
5
7
8
7
4
5
6
4
6
156. To do so we will rearrange the data like so:
We will subtract each of these
values from the grand mean,
square the result and sum them
all up.
Here is how we compute the
Grand Mean =
Football
Players
Ben
Bob
Bud
Bubba
Burt
Ben
Bob
Bud
Bubba
Burt
Ben
Bob
Bud
Bubba
Burt
Season
Before
Before
Before
Before
Before
During
During
During
During
During
After
After
After
After
After
Slices of
Pizza
5
7
8
9
10
4
5
7
8
7
4
5
6
4
6
157. To do so we will rearrange the data like so:
We will subtract each of these
values from the grand mean,
square the result and sum them
all up.
Here is how we compute the
Grand Mean =
Football
Players
Ben
Bob
Bud
Bubba
Burt
Ben
Bob
Bud
Bubba
Burt
Ben
Bob
Bud
Bubba
Burt
Season
Before
Before
Before
Before
Before
During
During
During
During
During
After
After
After
After
After
Slices of
Pizza
5
7
8
9
10
4
5
7
8
7
4
5
6
4
6
Pizza Slices Consumed
Football
Players
Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
158. To do so we will rearrange the data like so:
We will subtract each of these
values from the grand mean,
square the result and sum them
all up.
Here is how we compute the
Grand Mean =
Football
Players
Ben
Bob
Bud
Bubba
Burt
Ben
Bob
Bud
Bubba
Burt
Ben
Bob
Bud
Bubba
Burt
Season
Before
Before
Before
Before
Before
During
During
During
During
During
After
After
After
After
After
Slices of
Pizza
5
7
8
9
10
4
5
7
8
7
4
5
6
4
6
Pizza Slices Consumed
Football
Players
Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
Average of All
Observations =
6.3
159. To do so we will rearrange the data like so:
We will subtract each of these
values from the grand mean,
square the result and sum them
all up.
푆푆푡표푡푎푙 = Σ(푋푖푗 − 푋 )2
Football
Players
Ben
Bob
Bud
Bubba
Burt
Ben
Bob
Bud
Bubba
Burt
Ben
Bob
Bud
Bubba
Burt
Season
Before
Before
Before
Before
Before
During
During
During
During
During
After
After
After
After
After
Slices of
Pizza
5
7
8
9
10
4
5
7
8
7
4
5
6
4
6
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
160. To do so we will rearrange the data like so:
We will subtract each of these
values from the grand mean,
square the result and sum them
all up.
Football
Players
Ben
Bob
Bud
Bubba
Burt
Ben
Bob
Bud
Bubba
Burt
Ben
Bob
Bud
Bubba
Burt
Season
Before
Before
Before
Before
Before
During
During
During
During
During
After
After
After
After
After
Slices of
Pizza
5
7
8
9
10
4
5
7
8
7
4
5
6
4
6
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
푆푆푡표푡푎푙 = Σ(푋푖푗 − 푋 )2
161. To do so we will rearrange the data like so:
We will subtract each of
these values from the
grand mean, square the
result and sum them all
up.
푆푆푡표푡푎푙 = Σ(푋푖푗 − 푋 )2
Bob
Bob
Before
During
7
5
-
-
Football
Players
Season
Slices of
Pizza
Grand
Mean
Ben
Before
5
-
6.3
Bob Before 7 - 6.3
Bud
Before
8
-
6.3
Bubba
Before
9
-
6.3
Burt
Before
10
-
6.3
Ben
During
4
-
6.3
Bob During 5 - 6.3
Bud
During
7
-
6.3
Bubba
During
8
-
6.3
Burt
During
7
-
6.3
Ben
After
4
-
6.3
Bob
After
5
-
6.3
Bud
After
6
-
6.3
Bubba
After
4
-
6.3
Burt
After
6
-
6.3
162. To do so we will rearrange the data like so:
We will subtract each
of these values from
the grand mean,
square the result and
sum them all up.
Football
Players
Ben
Bob
Bud
Bubba
Burt
Ben
Bob
Bud
Bubba
Burt
Ben
Bob
Bud
Bubba
Burt
Season
Before
Before
Before
Before
Before
During
During
During
During
During
After
After
After
After
After
Slices of
Pizza
5
7
8
9
10
4
5
7
8
7
4
5
6
4
6
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Football
Players
Season Slices of
Pizza
Grand
Mean
Ben Before 5 - 6.3
=
Bob Before 7 - 6.3
=
Bud Before 8 - 6.3
=
Bubba Before 9 - 6.3
=
Burt Before 10 - 6.3
=
Ben During 4 - 6.3
=
Bob During 5 - 6.3
=
Bud During 7 - 6.3
=
Bubba During 8 - 6.3
=
Burt During 7 - 6.3
=
Ben After 4 - 6.3
=
Bob After 5 - 6.3
=
Bud After 6 - 6.3
=
Bubba After 4 - 6.3
=
Burt After 6 - 6.3
=
163. To do so we will rearrange the data like so:
We will subtract each
of these values from
the grand mean,
square the result and
sum them all up.
Football
Players
Football
Players
Ben
Bob
Bud
Ben Before 5 - 6.3 = -1.3
Bob Before 7 - 6.3 = 0.7
Bud Before 8 - 6.3 = 1.7
Bubba
Burt
Bubba Before 9 - 6.3 = 2.7
Burt Before 10 - 6.3 = 3.7
Ben
Bob
Bud
Ben During 4 - 6.3 = -2.3
Bob During 5 - 6.3 = -1.3
Bud During 7 - 6.3 = 0.7
Bubba
Burt
Bubba During 8 - 6.3 = 1.7
Burt During 7 - 6.3 = 0.7
Ben
Bob
Bud
Bubba
Burt
Season Slices
Season
of Pizza
Before
Before
Before
Before
Before
During
During
During
During
During
After
After
After
After
After
Slices of
Pizza
5
7
8
9
10
4
5
7
8
7
4
5
6
4
6
Grand
Mean
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Football
Players
Season Slices of
Pizza
Grand
Mean
Deviation
Ben Before 5 - 6.3
=
Bob Before 7 - 6.3
=
Bud Before 8 - 6.3
=
Bubba Before 9 - 6.3
=
Burt Before 10 - 6.3
=
Ben During 4 - 6.3
=
Bob During 5 - 6.3
=
Bud During 7 - 6.3
=
Bubba During 8 - 6.3
=
Burt During 7 - 6.3
=
Ben After 4 - 6.3
=
Bob After 5 - 6.3
=
Bud After 6 - 6.3
=
Bubba After 4 - 6.3
=
Burt After 6 - 6.3
=
Ben After 4 - 6.3 = -2.3
Bob After 5 - 6.3 = -1.3
Bud After 6 - 6.3 = -0.3
Bubba After 4 - 6.3 = -2.3
Burt After 6 - 6.3 = -0.3
164. To do so we will rearrange the data like so:
Football
Players
Season Slices of
Pizza
Grand
Mean
Deviation Squared
Ben Before 5 - 6.3 = -1.3 1.8
Bob Before 7 - 6.3 = 0.7 0.4
Bud Before 8 - 6.3 = 1.7 2.8
Bubba Before 9 - 6.3 = 2.7 7.1
Burt Before 10 - 6.3 = 3.7 13.4
Ben During 4 - 6.3 = -2.3 5.4
Bob During 5 - 6.3 = -1.3 1.8
Bud During 7 - 6.3 = 0.7 0.4
Bubba During 8 - 6.3 = 1.7 2.8
Burt During 7 - 6.3 = 0.7 0.4
Ben After 4 - 6.3 = -2.3 5.4
Bob After 5 - 6.3 = -1.3 1.8
Bud After 6 - 6.3 = -0.3 0.1
Bubba After 4 - 6.3 = -2.3 5.4
Burt After 6 - 6.3 = -0.3 0.1
We will subtract each of these values from the
grand mean, square the result and sum them all
up.
165. To do so we will rearrange the data like so:
Football
Players
Season Slices of
Pizza
Grand
Mean
Deviation Squared
Ben Before 5 - 6.3 = -1.3 1.8
Bob Before 7 - 6.3 = 0.7 0.4
Bud Before 8 - 6.3 = 1.7 2.8
Bubba Before 9 - 6.3 = 2.7 7.1
Burt Before 10 - 6.3 = 3.7 13.4
Ben During 4 - 6.3 = -2.3 5.4
Bob During 5 - 6.3 = -1.3 1.8
Bud During 7 - 6.3 = 0.7 0.4
Bubba During 8 - 6.3 = 1.7 2.8
Burt During 7 - 6.3 = 0.7 0.4
Ben After 4 - 6.3 = -2.3 5.4
Bob After 5 - 6.3 = -1.3 1.8
Bud After 6 - 6.3 = -0.3 0.1
Bubba After 4 - 6.3 = -2.3 5.4
Burt After 6 - 6.3 = -0.3 0.1
= 49.3
We will subtract each of these values from the
grand mean, square the result and sum them all
up.
166. To do so we will rearrange the data like so:
Football
Players
Season Slices of
Then –
Pizza
Grand
Mean
Deviation Squared
Ben Before 5 - 6.3 = -1.3 1.8
Bob Before 7 - 6.3 = 0.7 0.4
Bud Before 8 - 6.3 = 1.7 2.8
Bubba Before 9 - 6.3 = 2.7 7.1
Burt Before 10 - 6.3 = 3.7 13.4
Ben During 4 - 6.3 = -2.3 5.4
Bob During 5 - 6.3 = -1.3 1.8
Bud During 7 - 6.3 = 0.7 0.4
Bubba During 8 - 6.3 = 1.7 2.8
Burt During 7 - 6.3 = 0.7 0.4
Ben After 4 - 6.3 = -2.3 5.4
Bob After 5 - 6.3 = -1.3 1.8
Bud After 6 - 6.3 = -0.3 0.1
Bubba After 4 - 6.3 = -2.3 5.4
Burt After 6 - 6.3 = -0.3 0.1
= 49.3
167. To do so we will rearrange the data like so:
Football
Players
Season Slices of
Pizza
Grand
Mean
Deviation Squared
Ben Before 5 - 6.3 = -1.3 1.8
Bob Before 7 - 6.3 = 0.7 0.4
Bud Before 8 - 6.3 = 1.7 2.8
Bubba Before 9 - 6.3 = 2.7 7.1
Burt Before 10 - 6.3 = 3.7 13.4
Ben During 4 - 6.3 = -2.3 5.4
Bob During 5 - 6.3 = -1.3 1.8
Bud During 7 - 6.3 = 0.7 0.4
Bubba During 8 - 6.3 = 1.7 2.8
Burt During 7 - 6.3 = 0.7 0.4
Ben After 4 - 6.3 = -2.3 5.4
Bob After 5 - 6.3 = -1.3 1.8
Bud After 6 - 6.3 = -0.3 0.1
Bubba After 4 - 6.3 = -2.3 5.4
Burt After 6 - 6.3 = -0.3 0.1
= 49.3
Then – we place the total sums of squares result
in the ANOVA table.
168. To do so we will rearrange the data like so:
Football
Players
Season Slices of
Pizza
Grand
Mean
Deviation Squared
Ben Before 5 - 6.3 = -1.3 1.8
Bob Before 7 - 6.3 = 0.7 0.4
Bud Before 8 - 6.3 = 1.7 2.8
Bubba Before 9 - 6.3 = 2.7 7.1
Burt Before 10 - 6.3 = 3.7 13.4
Ben During 4 - 6.3 = -2.3 5.4
Bob During 5 - 6.3 = -1.3 1.8
Bud During 7 - 6.3 = 0.7 0.4
Bubba During 8 - 6.3 = 1.7 2.8
Burt During 7 - 6.3 = 0.7 0.4
Ben After 4 - 6.3 = -2.3 5.4
Bob After 5 - 6.3 = -1.3 1.8
Bud After 6 - 6.3 = -0.3 0.1
Bubba After 4 - 6.3 = -2.3 5.4
Burt After 6 - 6.3 = -0.3 0.1
= 49.3
Then – we place the total sums of squares result
in the ANOVA table.
169. Then – we place the total sums of squares result
in the ANOVA table.
Football
Players
Season Slices of
Pizza
Grand
Mean
Deviation Squared
Ben Before 5 - 6.3 = -1.3 1.8
Bob Before 7 - 6.3 = 0.7 0.4
Bud Before 8 - 6.3 = 1.7 2.8
Bubba Before 9 - 6.3 = 2.7 7.1
Burt Before 10 - 6.3 = 3.7 13.4
Ben During 4 - 6.3 = -2.3 5.4
Bob During 5 - 6.3 = -1.3 1.8
Bud During 7 - 6.3 = 0.7 0.4
Bubba During 8 - 6.3 = 1.7 2.8
Burt During 7 - 6.3 = 0.7 0.4
Ben After 4 - 6.3 = -2.3 5.4
Bob After 5 - 6.3 = -1.3 1.8
Bud After 6 - 6.3 = -0.3 0.1
Bubba After 4 - 6.3 = -2.3 5.4
Burt After 6 - 6.3 = -0.3 0.1
= 49.3
170. Then – we place the total sums of squares result
in the ANOVA table.
Football
Players
Season Slices of
Pizza
Grand
Mean
Deviation Squared
Ben Before 5 - 6.3 = -1.3 1.8
Bob Before 7 - 6.3 = 0.7 0.4
Bud Before 8 - 6.3 = 1.7 2.8
Bubba Before 9 - 6.3 = 2.7 7.1
Burt Before 10 - 6.3 = 3.7 13.4
Ben During 4 - 6.3 = -2.3 5.4
Bob During 5 - 6.3 = -1.3 1.8
Bud During 7 - 6.3 = 0.7 0.4
Bubba During 8 - 6.3 = 1.7 2.8
Burt During 7 - 6.3 = 0.7 0.4
Ben After 4 - 6.3 = -2.3 5.4
Bob After 5 - 6.3 = -1.3 1.8
Bud After 6 - 6.3 = -0.3 0.1
Bubba After 4 - 6.3 = -2.3 5.4
Burt After 6 - 6.3 = -0.3 0.1
= 49.3 Tests of Within-Subjects Effects
Measure: Pizza slices consumed
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
171. We have now calculated the total sums of
squares. This is a good starting point. Because
now we want to know of that total sums of
squares how many sums of squares are
generated from the following sources:
172. We have now calculated the total sums of
squares. This is a good starting point. Because
now we want to know of that total sums of
squares how many sums of squares are
generated from the following sources:
• Between subjects (this is the variance we
want to eliminate)
173. We have now calculated the total sums of
squares. This is a good starting point. Because
now we want to know of that total sums of
squares how many sums of squares are
generated from the following sources:
• Between subjects (this is the variance we
want to eliminate)
• Between Groups (this would be between
BEFORE, DURING, AFTER)
174. We have now calculated the total sums of
squares. This is a good starting point. Because
now we want to know of that total sums of
squares how many sums of squares are
generated from the following sources:
• Between subjects (this is the variance we
want to eliminate)
• Between Groups (this would be between
BEFORE, DURING, AFTER)
• Error (the variance that we cannot explain
with our design)
175. With these sums of squares we will be able to
compute our F ratio value and then statistical
significance.
176. With these sums of squares we will be able to
compute our F ratio value and then statistical
significance.
Tests of Within-Subjects Effects
Measure: Pizza slices consumed
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
177. With these sums of squares we will be able to
compute our F ratio value and then statistical
significance.
Tests of Within-Subjects Effects
Measure: Pizza slices consumed
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
Let’s calculate the sums of squares between
subjects.
178. Remember if we were just computing a one way
ANOVA the table would go from this:
179. Remember if we were just computing a one way
ANOVA the table would go from this:
Tests of Within-Subjects Effects
Measure: Pizza slices consumed
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
180. Remember if we were just computing a one way
ANOVA the table would go from this:
To this:
Tests of Within-Subjects Effects
Measure: Pizza slices consumed
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
181. Remember if we were just computing a one way
ANOVA the table would go from this:
To this:
Tests of Within-Subjects Effects
Measure: Pizza slices consumed
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
Tests of Within-Subjects Effects
Measure: Pizza slices consumed
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Groups 19.733 2 9.867 2.669 .078
Error 29.600 8 3.700
Total 49.333 14
182. Remember if we were just computing a one way
ANOVA the table would go from this:
To this:
Tests of Within-Subjects Effects
Measure: Pizza slices consumed
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
Tests of Within-Subjects Effects
Measure: Pizza slices consumed
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Groups 19.733 2 9.867 2.669 .078
Error 29.600 8 3.700
Total 49.333 14
183. All of that variability goes into the error or
within groups sums of squares (29.600) which
makes the F statistic smaller (from 9.548 to
2.669), the significance value no longer
significant (.008 to .078).
184. All of that variability goes into the error or
within groups sums of squares (29.600) which
makes the F statistic smaller (from 9.548 to
2.669), the significance value no longer
significant (.008 to .078).
But the difference in within groups variability is
not a function of error, it is a function of Ben,
Bob, Bud, Bubba, and Burt’s being different in
terms of the amount of slices they eat regardless
of when they eat!
185. All of that variability goes into the error or
within groups sums of squares (29.600) which
makes the F statistic smaller (from 9.548 to
2.669), the significance value no longer
significant (.008 to .078).
But the difference in within groups variability is
not a function of error, it is a function of Ben,
Bob, Bud, Bubba, and Burt’s being different in
terms of the amount of Pizza slices Slices Consumed
they eat regardless
Football
Before the
During the
After the
Average
of when they eat!
Players
Season
Season
Season
Ben 5 4 4 4.3
Bob 7 5 5 5.7
Bud 8 7 6 7.0
Bubba 9 8 4 7.0
Burt 10 7 6 7.7
186. Here is a data set where there are not between
group differences, but there is a lot of difference
based on when the group eats their pizza:
187. Here is a data set where there are not between
group differences, but there is a lot of difference
based on when the group eats their pizza:
Pizza Slices Consumed
Football
Players
Before the
Season
During the
Season
After the
Season
Average
Ben 1 5 9 5.0
Bob 2 5 8 5.0
Bud 3 5 7 5.0
Bubba 1 5 9 5.0
Burt 2 5 8 5.0
188. Here is a data set where there are not between
group differences, but there is a lot of difference
based on when the group eats their pizza:
Pizza Slices Consumed
Football
Players
Before the
Season
During the
Season
After the
Season
Average
Ben 1 5 9 5.0
Bob 2 5 8 5.0
Bud 3 5 7 5.0
Bubba 1 5 9 5.0
Burt 2 5 8 5.0
There is no variability between subjects (they
are all 5.0).
190. Look at the variability between groups:
Pizza Slices Consumed
Football
Players
Before the
Season
During the
Season
After the
Season
Average
Ben 1 5 9 5.0
Bob 2 5 8 5.0
Bud 3 5 7 5.0
Bubba 1 5 9 5.0
Burt 2 5 8 5.0
1.8 5.0 8.2
191. Look at the variability between groups:
Pizza Slices Consumed
Football
Players
Before the
Season
During the
Season
After the
Season
Average
Ben 1 5 9 5.0
Bob 2 5 8 5.0
Bud 3 5 7 5.0
Bubba 1 5 9 5.0
Burt 2 5 8 5.0
1.8 5.0 8.2
They are very different from one another.
193. Here is what the ANOVA table would look like:
Tests of Within-Subjects Effects
Measure: Pizza slices consumed
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 0.000 4
Between Groups 102.400 2 51.200 73.143 .000
Error 5.600 8 0.700
Total 49.333 14
194. Here is what the ANOVA table would look like:
Tests of Within-Subjects Effects
Measure: Pizza slices consumed
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 0.000 4
Between Groups 102.400 2 51.200 73.143 .000
Error 5.600 8 0.700
Total 49.333 14
Notice how there are no sum of squares values
for the between subjects source of variability!
195. Here is what the ANOVA table would look like:
Tests of Within-Subjects Effects
Measure: Pizza slices consumed
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 0.000 4
Between Groups 102.400 2 51.200 73.143 .000
Error 5.600 8 0.700
Total 49.333 14
Notice how there are no sum of squares values
for the between subjects source of variability!
But there is a lot of sum of squares values for
the between groups.
196. Here is what the ANOVA table would look like:
Tests of Within-Subjects Effects
Measure: Pizza slices consumed
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 0.000 4
Between Groups 102.400 2 51.200 73.143 .000
Error 5.600 8 0.700
Total 49.333 14
Notice how there are no sum of squares values
for the between subjects source of variability!
But there is a lot of sum of squares values for
the between groups.
Tests of Within-Subjects Effects
Measure: Pizza slices consumed
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 0.000 4
Between Groups 102.400 2 51.200 73.143 .000
Error 5.600 8 0.700
Total 49.333 14
197. What would the data set look like if there was
very little between groups (by season) variability
and a great deal of between subjects variability:
198. What would the data set look like if there was
very little between groups (by season) variability
and a great deal of between subjects variability:
Here it is:
199. What would the data set look like if there was
very little between groups (by season) variability
and a great deal of between subjects variability:
Here it is:
Pizza Slices Consumed
Football
Players
Before the
Season
During the
Season
After the
Season
Average
Ben 3 3 3 3.0
Bob 5 5 5 5.0
Bud 7 7 7 7.0
Bubba 8 8 8 8.0
Burt 12 12 13 12.3
Between
Subjects
200. In this case the between subjects (Ben, Bob, Bud
. . .), are very different.
201. In this case the between subjects (Ben, Bob, Bud
. . .), are very different.
When you see between SUBJECTS averages that
far away, you know that the sums of squares for
between groups will be very large.
202. In this case the between subjects (Ben, Bob, Bud
. . .), are very different.
When you see between SUBJECTS averages that
far away, you know that the sums of squares for
between groups will be very large.
Tests of Within-Subjects Effects
Measure: Pizza slices consumed
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 148.267 4
Between Groups 0.133 2 0.067 1.000 .689
Error 0.533 8 0.067
Total 148.933 14
203. Notice, in contrast, as we compute the between
group (seasons) average how close they are.
204. Notice, in contrast, as we compute the between
group (seasons) average how close they are.
Pizza Slices Consumed
Football
Players
Before the
Season
During the
Season
After the
Season
Average
Ben 3 3 3 3.0
Bob 5 5 5 5.0
Bud 7 7 7 7.0
Bubba 8 8 8 8.0
Burt 12 12 13 12.3
7.0 7.0 7.2
205. Notice, in contrast, as we compute the between
group (seasons) average how close they are.
Pizza Slices Consumed
Football
Players
Before the
Season
During the
Season
After the
Season
Average
Ben 3 3 3 3.0
Bob 5 5 5 5.0
Bud 7 7 7 7.0
Bubba 8 8 8 8.0
Burt 12 12 13 12.3
7.0 7.0 7.2
Between
Groups
206. Notice, in contrast, as we compute the between
group (seasons) average how close they are.
Pizza Slices Consumed
Football
Players
Before the
Season
During the
Season
After the
Season
Average
Ben 3 3 3 3.0
Bob 5 5 5 5.0
Bud 7 7 7 7.0
Bubba 8 8 8 8.0
Burt 12 12 13 12.3
7.0 7.0 7.2
Between
Groups
207. When you see between group averages this
close you know that the sums of squares for
between groups will be very small.
208. When you see between group averages this
close you know that the sums of squares for
between groups will be very small.
Tests of Within-Subjects Effects
Measure: Pizza slices consumed
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 148.267 4
Between Groups 0.133 2 0.067 1.000 .689
Error 0.533 8 0.067
Total 148.933 14
209. When you see between group averages this
close you know that the sums of squares for
between groups will be very small.
Tests of Within-Subjects Effects
Measure: Pizza slices consumed
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 148.267 4
Between Groups 0.133 2 0.067 1.000 .689
Error 0.533 8 0.067
Total 148.933 14
Now that we have conceptually considered the
sources of variability as described by the sum of
squares, let’s begin calculating between
subjects, between groups, and the error
sources.
210. We will begin with calculating Between Subjects
sum of squares.
211. We will begin with calculating Between Subjects
sum of squares.
To do so, let’s return to our original data set:
212. We will begin with calculating Between Subjects
sum of squares.
To do so, let’s return to our original data set:
Pizza Slices Consumed
Football
Players
Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
213. We will begin with calculating Between Subjects
sum of squares.
To do so, let’s return to our original data set:
Pizza Slices Consumed
Football
Players
Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
Here is the formula for calculating SS between
subjects.
214. We will begin with calculating Between Subjects
sum of squares.
To do so, let’s return to our original data set:
Pizza Slices Consumed
Football
Players
Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
Here is the formula for calculating SS between
subjects.
푆푆푏푒푡푤푒푒푛 푠푢푏푗푒푐푡푠 = 푘 ∗ Σ(푋푏푠 − 푋 )2
216. 푆푆푏푒푡푤푒푒푛 푠푢푏푗푒푐푡푠 = 푘 ∗ Σ(푿풃풔 − 푋 )2
Pizza Slices Consumed
Football
Players
Before the
Season
During the
Season
After the
Season
Average
Ben 5 4 4 4.3
Bob 7 5 5 5.7
Bud 8 7 6 7.0
Bubba 9 8 4 7.0
Burt 10 7 6 7.7
217. 푆푆푏푒푡푤푒푒푛 푠푢푏푗푒푐푡푠 = 푘 ∗ Σ(푿풃풔 − 푋 )2
This means the
average of between
Pizza Slices Consumed
Football
Players
Before the
Season
During the
Season
After the
Season
Average
subjects
Ben 5 4 4 4.3
Bob 7 5 5 5.7
Bud 8 7 6 7.0
Bubba 9 8 4 7.0
Burt 10 7 6 7.7
218. 푆푆푏푒푡푤푒푒푛 푠푢푏푗푒푐푡푠 = 푘 ∗ Σ(푿풃풔 − 푋 )2
Pizza Slices Consumed
Football
Players
Before the
Season
During the
Season
After the
Season
Average minus
Ben 5 4 4 4.3 -
Bob 7 5 5 5.7 -
Bud 8 7 6 7.0 -
Bubba 9 8 4 7.0 -
Burt 10 7 6 7.7 -
219. 푆푆푏푒푡푤푒푒푛 푠푢푏푗푒푐푡푠 = 푘 ∗ Σ(푿풃풔 − 푋 )2
This means the
average of all of
the observations
220. 푆푆푏푒푡푤푒푒푛 푠푢푏푗푒푐푡푠 = 푘 ∗ Σ(푿풃풔 − 푋 )2
Here is how we calculate the grand mean again:
221. 푆푆푏푒푡푤푒푒푛 푠푢푏푗푒푐푡푠 = 푘 ∗ Σ(푿풃풔 − 푋 )2
Here is how we calculate the grand mean again:
Pizza Slices Consumed
Football
Players
Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
Average of All
Observations =
6.3
222. 푆푆푏푒푡푤푒푒푛 푠푢푏푗푒푐푡푠 = 푘 ∗ Σ(푿풃풔 − 푋 )2
Here is how we calculate the grand mean again:
Now we subtract each subject or person average
from the Grand Mean.
223. 푆푆푏푒푡푤푒푒푛 푠푢푏푗푒푐푡푠 = 푘 ∗ Σ(푿풃풔 − 푋 )2
Here is how we calculate the grand mean again:
Now we subtract each subject or person average
from the Grand Mean.
Pizza Slices Consumed
Football
Players
Before the
Season
During the
Season
After the
Season
Average minus Grand
Mean
Ben 5 4 4 4.3 - 6.3
Bob 7 5 5 5.7 - 6.3
Bud 8 7 6 7.0 - 6.3
Bubba 9 8 4 7.0 - 6.3
Burt 10 7 6 7.7 - 6.3
This means the
average of all of
the observations
224. 푆푆푏푒푡푤푒푒푛 푠푢푏푗푒푐푡푠 = 푘 ∗ Σ(푿풃풔 − 푋 )2
Here is how we calculate the grand mean again:
Now we subtract each subject or person average
from the Grand Mean.
This gives us the person’s average score
deviation from the total or grand mean.
225. 푆푆푏푒푡푤푒푒푛 푠푢푏푗푒푐푡푠 = 푘 ∗ Σ(푿풃풔 − 푋 )2
Here is how we calculate the grand mean again:
Now we subtract each subject or person average
from the Grand Mean.
This gives us the person’s average score
deviation from the total or grand mean. Pizza Slices Consumed
Football
Players
Before the
Season
During the
Season
After the
Season
Average minus Grand
Mean
Deviation
Ben 5 4 4 4.3 - 6.3 -2.0
Bob 7 5 5 5.7 - 6.3 -0.6
Bud 8 7 6 7.0 - 6.3 0.7
Bubba 9 8 4 7.0 - 6.3 0.7
Burt 10 7 6 7.7 - 6.3 1.4
226. 푆푆푏푒푡푤푒푒푛 푠푢푏푗푒푐푡푠 = 푘 ∗ Σ(푿풃풔 − 푋 )2
Here is how we calculate the grand mean again:
Now we subtract each subject or person average
from the Grand Mean.
This gives us the person’s average score
deviation from the total or grand mean.
Now we will square the deviations.
227. 푆푆푏푒푡푤푒푒푛 푠푢푏푗푒푐푡푠 = 푘 ∗ Σ(푿풃풔 − 푋 )2
Here is how we calculate the grand mean again:
Now we subtract each subject or person average
from the Grand Mean.
This gives us the person’s average score
deviation from the total or grand mean.
Now we will square the deviations.
228. 푆푆푏푒푡푤푒푒푛 푠푢푏푗푒푐푡푠 = 푘 ∗ Σ(푿풃풔 − 푋 )2
Here is how we calculate the grand mean again:
Now we subtract each subject or person average
from the Grand Mean.
This gives us the person’s average score
deviation Pizza Slices from Consumed
the total or grand mean.
Football
Before
During the
After the
Average minus Grand
Deviation Squared
Players
the
Season
Season
Mean
Now Season
we will square the deviations
Ben 5 4 4 4.3 - 6.3 -2.0 3.9
Bob 7 5 5 5.7 - 6.3 -0.6 0.4
Bud 8 7 6 7.0 - 6.3 0.7 0.5
Bubba 9 8 4 7.0 - 6.3 0.7 0.5
Burt 10 7 6 7.7 - 6.3 1.4 1.9
229. 푆푆푏푒푡푤푒푒푛 푠푢푏푗푒푐푡푠 = 푘 ∗ Σ(푿풃풔 − 푋 )2
Here is how we calculate the grand mean again:
Now we subtract each subject or person average
from the Grand Mean.
This gives us the person’s average score
deviation from the total or grand mean.
Now we will square the deviations.
Then we sum all of these squared deviations.
230. 푆푆푏푒푡푤푒푒푛 푠푢푏푗푒푐푡푠 = 푘 ∗ Σ(푿풃풔 − 푋 )2
Here is how we calculate the grand mean again:
Now we subtract each subject or person average
from the Grand Mean.
This gives us the person’s average score
deviation from the total or grand mean.
Now we will square the deviations.
Then we sum all of these squared deviations.
231. 푆푆푏푒푡푤푒푒푛 푠푢푏푗푒푐푡푠 = 푘 ∗ Σ(푿풃풔 − 푋 )2
Here is how we calculate the grand mean again:
Now we subtract each subject or person average
from the Grand Mean.
This gives Pizza Slices us Consumed
the person’s average score
deviation Football
Before
During from the
After the the
Average minus Grand
Deviation Squared
Players
the
Season
Season
total or Mean
grand mean.
Season
Now Ben we 5 will 4 square 4 4.3 the - deviations.
6.3 -2.0 3.9
Bob 7 5 5 5.7 - 6.3 -0.6 0.4
Then Bud we 8 sum 7 all 6 of these 7.0 - 6.3 squared 0.7 deviations.
0.5
Bubba 9 8 4 7.0 - 6.3 0.7 0.5
Burt 10 7 6 7.7 - 6.3 1.4 1.9
7.1
Sum
up
232. Here is how we calculate the grand mean again:
Now we subtract each subject or person average
from the Grand Mean.
This gives us the person’s average score
deviation from the total or grand mean.
Now we will square the deviations.
Then we sum all of these squared deviations.
Finally, we multiply the sum all of these squared
deviations by the number of groups:
233. 푆푆푏푒푡푤푒푒푛 푠푢푏푗푒푐푡푠 = 푘 ∗ Σ(푿풃풔 − 푋 )2
Pizza Slices Consumed
Football
Players
Before
the
Season
During the
Season
After the
Season
Average minus Grand
Mean
Deviation Squared
Ben 5 4 4 4.3 - 6.3 -2.0 3.9
Bob 7 5 5 5.7 - 6.3 -0.6 0.4
Bud 8 7 6 7.0 - 6.3 0.7 0.5
Bubba 9 8 4 7.0 - 6.3 0.7 0.5
Burt 10 7 6 7.7 - 6.3 1.4 1.9
7.1
Times 3 groups
Sum of Squares Between Subjects 21.3
234. 푆푆푏푒푡푤푒푒푛 푠푢푏푗푒푐푡푠 = 푘 ∗ Σ(푿풃풔 − 푋 )2
Pizza Slices Consumed
Football
Players
Before
the
Season
During the
Season
After the
Season
Average minus Grand
Mean
Deviation Squared
Ben 5 4 4 4.3 - 6.3 -2.0 3.9
Number of
conditions
Bob 7 5 5 5.7 - 6.3 -0.6 0.4
Bud 8 7 6 7.0 - 6.3 0.7 0.5
Bubba 9 8 4 7.0 - 6.3 0.7 0.5
Burt 10 7 6 7.7 - 6.3 1.4 1.9
7.1
Times 3 groups
Sum of Squares Between Subjects 21.3
235. 푆푆푏푒푡푤푒푒푛 푠푢푏푗푒푐푡푠 = 푘 ∗ Σ(푿풃풔 − 푋 )2
Pizza Slices Consumed
Football
Players
Before
the
Season
During the
Season
After the
Season
Average minus Grand
Mean
Deviation Squared
Ben 5 4 4 4.3 - 6.3 -2.0 3.9
Bob 7 5 5 5.7 - 6.3 -0.6 0.4
Bud 8 7 6 7.0 - 6.3 0.7 0.5
Bubba 9 8 4 7.0 - 6.3 0.7 0.5
Burt 10 7 6 7.7 - 6.3 1.4 1.9
7.1
Times 3 groups
Sum of Squares Between Subjects 21.3
236. 푆푆푏푒푡푤푒푒푛 푠푢푏푗푒푐푡푠 = 푘 ∗ Σ(푿풃풔 − 푋 )2
Pizza Slices Consumed
Football
Players
Before
the
Season
During the
Season
After the
Season
Average minus Grand
Mean
Deviation Squared
Ben 5 4 4 4.3 - 6.3 -2.0 3.9
Bob 7 5 5 5.7 - 6.3 -0.6 0.4
Bud 8 7 6 7.0 - 6.3 0.7 0.5
Bubba 9 8 4 7.0 - 6.3 0.7 0.5
Burt 10 7 6 7.7 - 6.3 1.4 1.9
7.1
Times 3 groups
Sum of Squares Between Subjects 21.3
237. 푆푆푏푒푡푤푒푒푛 푠푢푏푗푒푐푡푠 = 푘 ∗ Σ(푿풃풔 − 푋 )2
Pizza Slices Consumed
Football
Players
Before
the
Season
During the
Season
After the
Season
Average minus Grand
Mean
Deviation Squared
Ben 5 4 4 4.3 - 6.3 -2.0 3.9
Bob 7 5 5 5.7 - 6.3 -0.6 0.4
Bud 8 7 6 7.0 - 6.3 0.7 0.5
Bubba 9 8 4 7.0 - 6.3 0.7 0.5
Burt 10 7 6 7.7 - 6.3 1.4 1.9
7.1
Times 3 groups
Sum of Squares Between Subjects 21.3
1 2 3
238. 푆푆푏푒푡푤푒푒푛 푠푢푏푗푒푐푡푠 = 푘 ∗ Σ(푿풃풔 − 푋 )2
Pizza Slices Consumed
Football
Players
Before
the
Season
During the
Season
After the
Season
Average minus Grand
Mean
Deviation Squared
Ben 5 4 4 4.3 - 6.3 -2.0 3.9
Bob 7 5 5 5.7 - 6.3 -0.6 0.4
Bud 8 7 6 7.0 - 6.3 0.7 0.5
Bubba 9 8 4 7.0 - 6.3 0.7 0.5
Burt 10 7 6 7.7 - 6.3 1.4 1.9
7.1
Times 3 groups
Sum of Squares Between Subjects 21.3
1 2 3
239. 푆푆푏푒푡푤푒푒푛 푠푢푏푗푒푐푡푠 = 푘 ∗ Σ(푿풃풔 − 푋 )2
Pizza Slices Consumed
Football
Players
Before
the
Season
During the
Season
Tests of Within-Subjects Effects
Measure: Pizza slices consumed
Source
Type III Sum
of Squares df
After the
Season
Average minus Grand
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
Mean
Deviation Squared
Ben 5 4 4 4.3 - 6.3 -2.0 3.9
Bob 7 5 5 5.7 - 6.3 -0.6 0.4
Bud 8 7 6 7.0 - 6.3 0.7 0.5
Bubba 9 8 4 7.0 - 6.3 0.7 0.5
Burt 10 7 6 7.7 - 6.3 1.4 1.9
7.1
Times 3 groups
Sum of Squares Between Subjects 21.3
1 2 3
240. 푆푆푏푒푡푤푒푒푛 푠푢푏푗푒푐푡푠 = 푘 ∗ Σ(푿풃풔 − 푋 )2
Pizza Slices Consumed
Football
Players
Before
the
Season
During the
Season
After the
Season
Average minus Grand
Mean
Deviation Squared
Ben 5 4 4 4.3 - 6.3 -2.0 3.9
Bob 7 5 5 5.7 - 6.3 -0.6 0.4
Bud 8 7 6 7.0 - 6.3 0.7 0.5
Bubba 9 8 4 7.0 - 6.3 0.7 0.5
Burt 10 7 6 7.7 - 6.3 1.4 1.9
7.1
Times 3 groups
Sum of Squares Between Subjects 21.3
1 2 3
Tests of Within-Subjects Effects
Measure: Pizza slices consumed
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
241. Now it is time to compute the between groups
(seasons) sum of squares.
242. Now it is time to compute the between groups’
(seasons) sum of squares.
Here is the equation we will use to compute it:
243. Now it is time to compute the between groups’
(seasons) sum of squares.
Here is the equation we will use to compute it:
푛 ∗ Σ(푋 푘 − 푋 )
246. Let’s break this down with our data set:
푛 ∗ Σ(푋 푘 − 푋 )
Pizza Slices Consumed
Football
Players
Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
247. We begin by computing the mean of each
condition (k)
푛 ∗ Σ(푋 푘 − 푋 )
Pizza Slices Consumed
Football
Players
Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
248. We begin by computing the mean of each
condition (k)
푛 ∗ Σ(푋 푘 − 푋 )
Pizza Slices Consumed
Football
Players
Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
Condition
Mean
249. We begin by computing the mean of each
condition (k)
푛 ∗ Σ(푋 푘 − 푋 )
Pizza Slices Consumed
Football
Players
Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
Condition
Mean
7.8
250. We begin by computing the mean of each
condition (k)
푛 ∗ Σ(푋 푘 − 푋 )
Pizza Slices Consumed
Football
Players
Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
Condition
Mean
7.8
6.2
251. We begin by computing the mean of each
condition (k)
푛 ∗ Σ(푋 푘 − 푋 )
Pizza Slices Consumed
Football
Players
Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
Condition
Mean
7.8
6.2
5.0
252. Then subtract each condition mean from the
grand mean.
푛 ∗ Σ(푋 푘 − 푋 )
Pizza Slices Consumed
Football
Players
Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
Condition
Mean
7.8
6.2
5.0
253. Then subtract each condition mean from the
grand mean.
푛 ∗ Σ(푋 푘 − 푋 )
Pizza Slices Consumed
Football
Players
Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
Condition
Mean
7.8
6.2
5.0
minus - - -
254. Then subtract each condition mean from the
grand mean.
푛 ∗ Σ(푋 푘 − 푋 )
Pizza Slices Consumed
Football
Players
Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
Condition
Mean
7.8 6.2 5.0
minus - - -
Grand
Mean
6.3 6.3 6.3
255. Then subtract each condition mean from the
grand mean.
푛 ∗ Σ(푋 푘 − 푋 )
Pizza Slices Consumed
Football
Players
Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
Condition
Mean
7.8 6.2 5.0
minus - - -
Grand
Mean
6.3 6.3 6.3
equals
Deviation 1.5 -0.1 -1.3
256. Square the deviation.
푛 ∗ Σ(푋 푘 − 푋 )ퟐ
Pizza Slices Consumed
Football
Players
Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
Condition
Mean
7.8 6.2 5.0
minus - - -
Grand
Mean
6.3 6.3 6.3
equals
Deviation 1.5 -0.1 -1.3
Squared
Deviation
2.2 0.0 1.8
259. Sum the Squared Deviations: 푛 ∗ Σ(푋 푘 − 푋 )ퟐ
Pizza Slices Consumed
Football
Players
Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
Condition
Mean
7.8 6.2 5.0
minus - - -
Grand
Mean
6.3 6.3 6.3
equals
Deviation 1.5 -0.1 -1.3
Squared
Deviation
2.2 0.0 1.8
Sum
260. Football
Players
Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Sum the Squared Deviations: Bob 7 5 푛 ∗ 5
Σ(푋 푘 − 푋 )ퟐ
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
Condition
Mean
7.8 6.2 5.0
minus - - -
Grand
Mean
6.3 6.3 6.3
equals
Deviation 1.5 -0.1 -1.3
Squared
Deviation
2.2 0.0 1.8
Sum
3.95
Sum of Squared
Deviations
261. Multiply by the number of observations per
condition (number of pizza eating slices across
before, during, and after).
262. Multiply by the number of observations per
condition (number of pizza eating slices across
before, during, and after).
3.95
Sum of Squared
Deviations
263. Multiply by the number of observations per
condition (number of pizza eating slices across
before, during, and after).
3.95
Sum of Squared
Deviations
264. Multiply by the number of observations per
condition (number of pizza eating slices across
before, during, and after).
3.95
Sum of Squared
Deviations
5
Number of
observations
265. Multiply by the number of observations per
condition (number of pizza eating slices across
before, during, and after).
3.95
Sum of Squared
Deviations
5
Number of
observations
266. Multiply by the number of observations per
condition (number of pizza eating slices across
before, during, and after).
3.95
Sum of Squared
Deviations
5
Number of
observations
19.7
Weighted Sum of
Squared Deviations
267. Let’s return to the ANOVA table and put the
weighted sum of squared deviations.
268. Let’s return to the ANOVA table and put the
weighted sum of squared deviations.
Tests of Within-Subjects Effects
Measure: Pizza_slices
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
269. Let’s return to the ANOVA table and put the
weighted sum of squared deviations.
Tests of Within-Subjects Effects
Measure: Pizza_slices
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
3.95
Sum of Squared
Deviations
5
Number of
observations
19.7
Weighted Sum of
Squared Deviations
270. Let’s return to the ANOVA table and put the
weighted sum of squared deviations.
Tests of Within-Subjects Effects
Measure: Pizza_slices
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
3.95
Sum of Squared
Deviations
5
Number of
observations
19.7
Weighted Sum of
Squared Deviations
271. So far we have calculated Total Sum of Squares
along with Sum of Squares for Between
Subjects, and Between Groups.
272. So far we have calculated Total Sum of Squares
along with Sum of Squares along with Sum of
Squares for Between Subjects, Between Groups.
Tests of Within-Subjects Effects
Measure: Pizza_slices
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
273. Now we will calculate the sum of squares
associated with Error.
274. Now we will calculate the sum of squares
associated with Error.
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
275. To do this we simply add the between subjects
and between groups sums of squares.
276. To do this we simply add the between subjects
and between groups sums of squares.
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
277. To do this we simply add the between subjects
and between groups sums of squares.
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
21.333
Between Subjects
Sum of Squares
19.733
Between Groups
Sum of Squares
41.600
Between Subjects &
Groups Sum of
Squares Combined
278. Then we subtract the Between Subjects & Group
Sum of Squares Combined (41.600) from the
Total Sum of Squares (49.333)
279. Then we subtract the Between Subjects & Group
Sum of Squares Combined (41.600) from the
Total Sum of Squares (49.333)
49.333
Total Sum of Squares
41.600
Between Subjects &
Groups Sum of Squares
Combined
8.267
Sum of Squares
Attributed to Error
or Unexplained
280. Then we subtract the Between Subjects & Group
Sum of Squares Combined (41.600) from the
Total Sum of Squares (49.333)
49.333
Total Sum of Squares
41.600
Between Subjects &
Groups Sum of Squares
Combined
8.267
Sum of Squares
Attributed to Error
or Unexplained
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
281. Now we have all of the information necessary to
determine if there is a statistically significant
difference between pizza slices consumed by
football players between three different eating
occasions (before, during or after the season).
282. Now we have all of the information necessary to
determine if there is a statistically significant
difference between pizza slices consumed by
football players between three different eating
occasions (before, during or after the season).
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
284. To calculate the significance level
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
286. We must calculate the F ratio
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
287. Which is calculated by dividing the Between
Groups Mean Square value (9.867) by the Error
Mean Square value (1.033).
288. Which is calculated by dividing the Between
Groups Mean Square value (9.867) by the Error
Mean Square value (1.033).
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 =
9.548 .008
Error 8.267 8 1.033
Total 49.333 14
289. Which is calculated by dividing the sum of
squares between groups by its degrees of
freedom, as shown below:
290. Which is calculated by dividing the sum of
squares between groups by its degrees of
freedom, as shown below:
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 =
9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
291. Which is calculated by dividing the sum of
squares between groups by its degrees of
freedom, as shown below:
And
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 =
9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
292. Which is calculated by dividing the sum of
squares between groups by its degrees of
freedom, as shown below:
=
And
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 =
1.033
Total 49.333 14
293. Which is calculated by dividing the sum of
squares between groups by its degrees of
freedom, as shown below:
Type III Sum
of Squares df
Between Subjects 21.333 4
Between Groups 19.733 2 =
9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
And
Source
Mean
Square F Sig.
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 =
1.033
Total 49.333 14
Now we need to figure out how we calculate
degrees of freedom for each source of sums of
squares.
294. Let’s begin with determining the degrees of
freedom Between Subjects.
295. Let’s begin with determining the degrees of
freedom Between Subjects.
296. Let’s begin with determining the degrees of
freedom Between Subjects.
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
297. Let’s begin with determining the degrees of
freedom Between Subjects.
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
We take the number of subjects which, in this
case, is 5 – 1 = 4
298. Let’s begin with determining the degrees of
freedom Between Subjects.
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
We take the number of subjects which, in this
case, is 5 – 1 = 4
299. Let’s begin with determining the degrees of
freedom Between Subjects.
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
We take the number of subjects which, in this
case, is 5 – 1 = 4
Pizza Slices Consumed
Football
Players
Before the
Season
During the
Season
After the
Season
Average
Ben 3 3 3 3.0
Bob 5 5 5 5.0
Bud 7 7 7 7.0
Bubba 8 8 8 8.0
Burt 12 12 13 12.3
Between
Subjects
1
2
3
4
5
300. Now – onto Between Groups Degrees of
Freedom (df)
301. Now – onto Between Groups Degrees of
Freedom (df)
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
302. Now – onto Between Groups Degrees of
Freedom (df)
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
We take the number of groups which in this case
is 3 – 1 = 2
303. Now – onto Between Groups Degrees of
Freedom (df)
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
We take the number of groups which in this case
is 3 – 1 = 2
304. Now – onto Between Groups Degrees of
Freedom (df)
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
We take the number of groups which in this case
is 3 – 1 = 2
1 2 3
Pizza Slices Consumed
Football
Players
Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
305. Now – onto Between Groups Degrees of
Freedom (df)
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
We take the number of groups which in this case
is 3 – 1 = 2
1 2 3
Pizza Slices Consumed
Football
Players
Before the
Season
During the
Season
After the
Season
Ben 5 4 4
Bob 7 5 5
Bud 8 7 6
Bubba 9 8 4
Burt 10 7 6
306. The error degrees of freedom are calculated by
multiplying the between subjects by the
between groups degrees of freedom.
307. The error degrees of freedom are calculated by
multiplying the between subjects by the
between groups degrees of freedom.
4
Between Subjects
Degrees of Freedom
2
Between Groups
Degrees of Freedom
8
Error Degrees of
Freedom
308. The error degrees of freedom are calculated by
multiplying the between subjects by the
between groups degrees of freedom.
4
Between Subjects
Degrees of Freedom
2
Between Groups
Degrees of Freedom
8
Error Degrees of
Freedom
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
309. The error degrees of freedom are calculated by
multiplying the between subjects by the
between groups degrees of freedom.
4
Between Subjects
Degrees of Freedom
2
Between Groups
Degrees of Freedom
8
Error Degrees of
Freedom
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
310. The degrees of freedom for total sum of squares
is calculated by adding all of the degrees of
freedom from the other three sources.
311. The degrees of freedom for total sum of squares
is calculated by adding all of the degrees of
freedom from the other three sources.
4 2 8 14
312. The degrees of freedom for total sum of squares
is calculated by adding all of the degrees of
freedom from the other three sources.
4 2 8 14
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
313. The degrees of freedom for total sum of squares
is calculated by adding all of the degrees of
freedom from the other three sources.
4 2 8 14
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
314. We will compute the Mean Square values for
just the Between Groups and Error. We are not
interested in what is happening with Between
Subjects. We calculated the Between Subjects
sum of squares only take out any differences
that are a function of differences that would
exist regardless of what group we were looking
at.
315. Once again, if we had not pulled out Between
Subjects sums of squares, then the Between
Subjects would be absorbed in the error value:
316. Once again, if we had not pulled out Between
Subjects sums of squares, then the Between
Subjects would be absorbed in the error value:
Tests of Within-Subjects Effects
Measure: Pizza_slices
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
317. Once again, if we had not pulled out Between
Subjects sums of squares, then the Between
Subjects would be absorbed in the error value:
Tests of Within-Subjects Effects
Measure: Pizza_slices
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Groups 19.733 2 9.867 4.000 .047
Within Groups 29.600 8 1.033
Total 49.333 14
318. Once again, if we had not pulled out Between
Subjects sums of squares, then the Between
Subjects would be absorbed in the error value:
Tests of Within-Subjects Effects
Measure: Pizza_slices
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Groups 19.733 2 9.867 4.000 .047
Within Groups 29.600 8 1.033
Total 49.333 14
319. Once again, if we had not pulled out Between
Subjects sums of squares, then the Between
Subjects would be absorbed in the error value:
Tests of Within-Subjects Effects
Measure: Pizza_slices
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Groups 19.733 2 9.867 4.000 .047
Within Groups 29.600 8 1.033
Total 49.333 14
Within Groups is
another way of
saying Error
320. And that would have created a larger error
mean square value:
321. And that would have created a larger error
mean square value:
Measure: Pizza_slices
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
Measure: Pizza_slices
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Groups 19.733 2 9.867 4.000 .047
Error 29.600 12 2.467
Total 49.333 14
322. And that would have created a larger error
mean square value:
Measure: Pizza_slices
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
Measure: Pizza_slices
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Groups 19.733 2 9.867 4.000 .047
Error 29.600 12 2.467
Total 49.333 14
323. Which in turn would have created a smaller F
value:
324. Which in turn would have created a smaller F
value:
Measure: Pizza_slices
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
Measure: Pizza_slices
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Groups 19.733 2 9.867 4.000 .047
Error 29.600 12 2.467
Total 49.333 14
325. Which in turn would have created a smaller F
value:
Measure: Pizza_slices
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
Measure: Pizza_slices
Source
Type III Sum
of Squares df
=
Mean
Square F Sig.
Between Groups 19.733 2 9.867 =
4.000 .047
Error 29.600 12 2.467
Total 49.333 14
326. Which in turn would have created a larger
significance value:
327. Which in turn would have created a larger
significance value:
Measure: Pizza_slices
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Subjects 21.333 4
Between Groups 19.733 2 9.867 9.548 .008
Error 8.267 8 1.033
Total 49.333 14
Measure: Pizza_slices
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Between Groups 19.733 2 9.867 4.000 .047
Error 29.600 12 2.467
Total 49.333 14