Planning the analysis and interpretation of resseaech data
1.
PLANNING THE ANALYSIS AND
INTERPRETATION OF
RESEARCH DATA
English 4a
A203
July 22, 2013
2.
o The design of a study does not only
consist of the procedures a researcher will
employ in the gathering of data but also
includes the researcher’s plan on how
collected data will be analyzed.
o It deals with the procedures in analyzing
both qualitative and quantitative data, as
well as the guidelines in choosing the
appropriate statistical techniques for
analyzing quantitative research data.
3.
TYPES OF DATA ANALYZED IN RESEARCH
Qualitative Data- mostly verbal.
Ex. Gender, socio economic status, religious preference
Quantitative Data- mostly numerical
Ex. Scores on achievement tests, numbers of hours of
study, or weight of a subject.
hahahaQualitative DataQuantitative Data
Overview.docx
Through analysis, a researcher can do the following
things;
1. Describe the data clearly;
2. Identify what is typical or atypical among the data;
3. Answer research questions or test hypothesis.
4.
Qualitative are analyzed logico-inductively;
1. Observations are made of behaviors,
situations, interactions, objects and
environments.
2. Topics are identified from the observations
and are scrutinized to discover patterns and
categories.
3. Conclusions are deduced from what is
observed and are stated verbally to answer
research questions.
5.
Quantitative Data are analyzed
mathematically and the results are
expressed in statistical terminology.
1. Depict what is typical and atypical among
the data;
2. Show degrees of difference or relationship
between two or more variables; and
3. Determine the likelihood that findings are
real for the population as opposed to having
occurred by chance in the sample.
7.
METHODS OF ANALYZING QUALITATIVE DATA
o Researcher has to present in greater details
the nature or characteristics of the
phenomenon or situation being described.
o Data analysis may take any of the following
forms: a) establishing categories or
typologies and determining the sequence of
events or patterns of behaviour.
8.
METHODS OF ANALYZING QUALITATIVE DATA
Historical Analysis- can be utilized when
the researcher is after explaining events or
phenomenon in the past so as to understand
the present.
- generalization in
historical analysis are arrived at, based on
the pattern of events that the researcher is
able to discover.
hahahaExamples of Historical Analysis.docx
9.
METHODS OF ANALYZING QUALITATIVE DATA
Inductive Analysis- this method of
analyzing qualitative data follows the pattern
of thinking and reasoning that starts from
specific to universal.
- the process starts from
particular observations and ends up with
generalization based on this specific
observations.
10.
METHODS OF ANALYZING QUALITATIVE DATA
Deductive Analysis- exactly opposite of the
inductive method of analysis.
- the researcher has to
begin with a general statement about a
phenomenon, situation or object and ends up
by providing details, particulars or specific
facts to support the said general statement.
11.
METHODS OF ANALYZING QUALITATIVE DATA
Content Analysis- is appropriate when the
researcher is concerned about explaining the
status of some at a particular time or its
development over a period of time using
available documents.
- is also called documentary
analysis.
- sources of data for this
method are records, reports, printed forms,
letters, autobiographies, diaries, books,
periodicals, films, cartoons etc.
hahahaHow to Do Content Analysis.docx
13.
METHODS IN ANALYZING QUANTITATIVE DATA
o Quantitative analysis is employed when the
data to be analyzed are numerical or
information which was assigned numerical
values to facilitate counting, summarization,
comparison and generalization.
- (Ardales, 1992)
o This type of analysis relies heavily on
statistical techniques.
o hahahaIn.docx
14.
METHODS IN ANALYZING QUANTITATIVE DATA
o Through statistics; researchers can-
1. Summarize data and reveal what is typical and
atypical within a group;
2. Show similarities and differences among
groups with the use of tests off differences;
3. Identify that is inherent in the selection of
samples;
4. Test for significance of findings; and
5. Make other inferences about the population.
16.
ANALYTIC PROCEDURES FOR QUANTITATIVE
DATA
Five Types of Analysis
Descriptive Analysis- the researcher is only
after describing the characteristics of the
subjects under study. Data are usually
analyzed to;
o Identify the general characteristics of a group , with the
use of descriptive statistics such as frequency,
percentage, mean, median mode and
o Determine the differences in the group or how members
of a group vary with reference to a given variable or factor
being studied with the use of standard deviation and
coefficient of variation.
hahahaProducing Descriptive Summary Data Using
SPSS.docx
17.
ANALYTIC PROCEDURES FOR QUANTITATIVE
DATA
Univariate Analysis- is utilized when the
researcher wants to analyze one variable or
factor at a time, such as levels of
commitment or job performance.
- relies heavily on the use
of the following summary statistics:
measures of central tendency; and
measures of variability.
hahahaFor example.docx
18.
ANALYTIC PROCEDURES FOR QUANTITATIVE
DATA
Measures of Central Tendency- to
co0mmunicate where scores or observations
center in the distribution.
1. Mean- is computed by dividing the sum of the
values by he number of cases.
2. Median- the middlemost value in an array,
such that 505 are below it and 50% are above
it.
3. Mode- is the category or value with the
greatest frequency of cases. It is the only
acceptable indicator of the most typical case
for data which are nominal or categorical.
19.
ANALYTIC PROCEDURES FOR QUANTITATIVE
DATA
Measures of Variability- reflect the amount of variation in the
score of distribution.
1. Minimum and Maximum Values- the minimum value
indicates how far the spread toward the lower direction and
the maximum value shows the extent of spread toward the
upper direction from the average.
2. Range- is the simply the distance and difference between
the maximum and minimum values, showing the total
spread between extremes.
3. Standard Deviation- measure of deviation that spread
away from the mean.
4. Quartile Deviation- is appropriate measure of variability to
employ when the median is the average used in describing
a given distribution.
20.
ANALYTIC PROCEDURES FOR QUANTITATIVE
DATA
Bivariate Analysis- this type analysis is used
when the researcher is interested in probing into
the relationship of two variables at a time.
Multivariate Analysis- is utilized when there
are researched questions which cannot be
responded using bivariate analysis.
- multiple regression analysis and
multiple classification analysis.
hahahaExamples of multivariate regression.docx
21.
ANALYTIC PROCEDURES FOR QUANTITATIVE
DATA
Comparative Analysis- when research
participants have to be compared on the
basis of certain variables being studied.
Example: a researcher who is after looking
into the differences in the work attitudes of
the rank-and-file and managerial employees
of one company, can used the
aforementioned analytic procedure.
hahaha1.jpg
22.
CHOOSING THE APPROPRIATE
STATISTICAL TESTS AND TECHNIQUES
23.
CHOOSING THE APPROPRIATE STATISTICAL
TESTS AND TECHNIQUES
Type of Research Questions-
“Three types of research questions usually
posed by an investigator: descriptive,
relationship, and difference.”
- Kumar. 1998
Nature of Raw Data- Diekhoff (1992)
categorized data into three types, namely:
categorical or nominal, ordinal, and metric.
- if nominal or ordinal,
non-parametric tests are appropriate to use; if
metric, parametric tests are deemed feasible to
apply.
24.
CHOOSING THE APPROPRIATE STATISTICAL
TESTS AND TECHNIQUES
3 Types of Data
Nominal Data- data or the number of
individuals or items falling under a particular
category or group.
Ex. When researcher records the number of
respondents according to gender or civil
status.
Ordinal Data- are data about rank or order.
25.
CHOOSING THE APPROPRIATE STATISTICAL
TESTS AND TECHNIQUES
Example of Likert Scale
26.
CHOOSING THE APPROPRIATE STATISTICAL
TESTS AND TECHNIQUES
Metric Data- data which can be subjected to
mathematical computations. That is can be added ,
subtracted, multiplied and divided.
Ex. Age, temperature reading, monetary transaction
and heights.
Hypothesis to be Tested- if the researcher is probing
into the association of two or more characteristics of
variables, he has to employ correlational statistics
and tests for determining the significance of the
computed correlation coefficient.
- Downing & Clark. 1997
27.
CHOOSING THE APPROPRIATE STATISTICAL
TESTS AND TECHNIQUES
Assumptions About the Nature of the
Population- used either parametric tests or
non-parametric tests of significance.
28.
SOME USEFUL PARAMETRIC AND NON-
PARAMETRIC TECHNIQUES
29.
SOME USEFUL PARAMETRIC AND NON-
PARAMETRIC TECHNIQUES
- “Most research in the academe is done in
one or two ways, either two or more groups
are compared or variables within one group
are related.”
- Frankel & Wallen. 1994
- Some of the most commonly used measures
of relationship and differences, as well as
their uses are presented below to guide in
preparing the statistical design of your
research proposal.
30.
TESTS OF REALTIONSHIP OR ASSOCIATION
Pearson-Product Moment Correlation (R)- is
calculated to show linear relationships between
two variables.
Spearman Ranks (rho)- used when ranks are
available for each of the two variables being
related.
Coefficient of Concordance (W)- usually
applied when the researcher wants to determine
whether agreements exist among the rankings
of three or more groups of respondents on a
particular variable under study.
31.
TESTS OF REALTIONSHIP OR ASSOCIATION
Chi-sauare Test (X2)- used as an inferential
statistics for nominal or categorical data. When
employed as test relationship, it is called test
independence. When used as a test difference,
it is considered a test of homogeneity.
Cramer’s V Statistics- used for assessing the
strength of the association between two
variables which were found to be significantly
related through the chi0square test of
independence .
32.
TESTS OF REALTIONSHIP OR ASSOCIATION
Point Biserial Correlation- used as a
measure of relationship between two
variable, where ne is continous and the other
is dichotomous.
Phi Coefficient- another measure of
relationship appropriate when two variables
correlated are both dichotomous.
Coefficient of Determination and
Alienation- these two measures have to be
computed when a significant R is obtained.
33.
TESTS OF REALTIONSHIP OR ASSOCIATION
Partial Correlation- is a correlational
method involving two or more variables.
Multiple Correlation- is a measure of
relationship appropriate when one dependent
variable is related to two or more
independent or predictor variables.
34.
TESTS OF DIFFERENCE
T test for Independent Samples- a parametric
test that is used in determining whether the
mean value of a variable in one group of
subjects is different from the mean value on the
same variable with the same group of subjects.
Fmax Hartley Test- is used in comparing the
standard deviations or variances of two or more
groups of research subjects on a variable being
studied.
Mann-Whitney U Test- is the non-parametric
counterpart of the independent T test.
35.
TESTS OF REALTIONSHIP OR ASSOCIATION
Sign Test- can also be used in determining
the significance of differences between two
sets of data from correlated samples.
Median Test- is a sign test for two
independent samples, in contrast to two
correlated samples.
Critical Ratio or Z test- often used in
determining the significance of differences
between two give percentages or
proportions, when they are not correlated.
36.
TESTS OF REALTIONSHIP OR ASSOCIATION
T test for Correlated Samples- is used
when two groups that have been matched
are being compared as in a pretest-posttest
design to see if any bserved mean gain is
significant.
Sandler’s A Test- is the non-parametric
analog of the T test for correlated samples.
Wilcoxon Rank Sum Test- is another non-
parametric alternative to difference of means
for correlated samples.
37.
TESTS OF REALTIONSHIP OR ASSOCIATION
Kolmoorov-Smirnov Test- this test fulfills the
function of the chi-square test in testing the
goodness-of-fit and the Wilcoxon Rank Sum
Test in determining whether the random
samples are from the same population.
Analysis of Variance (ANOVA)- used when the
researcher wants to find out if there are
significant differences between the means of
two or more groups on a variable under study.
Kruskall-Wallis H Test- this test looks for the
significance of differences among three or more
groups on a variable under study.
38.
TESTS OF REALTIONSHIP OR ASSOCIATION
Friedman Analysis of Variance (Fr)- is the
non-parametric analog of two-way ANOVA.
Analysis of Covariance (ANCOVA)- a
statistical technIque for equating groups in one
or more variables when testing for statistical
significance.
Multivatiate Analysis Of Variance (MANOVA)-
is an extension of ANOVA, which incorporates
two or more dependent variables in the same
analysis.
Clipping is a handy way to collect and organize the most important slides from a presentation. You can keep your great finds in clipboards organized around topics.
Be the first to comment