Analysis of data

         By:- PARTH
ď‚—   The analysis and interpretation of data represent
    the application of deductive and inductive logic to
    the research process.

ď‚—   The data are often classified by division into,
    subgroups, and are then analyzed and
    synthesized in such a way that hypothesis may
    be verified or rejected.

ď‚—   The final result may be a new principle or
    generalization. Data are examined in terms of
    comparison between the more homogeneous
    segments within the group any by comparison
    with some outside criteria.
ď‚—   NEED FOR ANALYSIS OF DATA OR
    TREATMENT OF DATA

ď‚—   After administering and scoring research tools
    scripts, data collected and organized. The
    collected data are known as „raw data.‟

ď‚—   The raw data are meaningless unless certain
    statistical treatment is given to them.

ď‚—   Analysis of data means to make the raw data
    meaningful or to draw some results from the
    data after the proper treatment.
ď‚—   Thus, the analysis of data serves the following
    main functions:


ď‚—   1. To make the raw data meaningful,
ď‚—   2. To test null hypothesis,
ď‚—   3. To obtain the significant results,
ď‚—   4. To draw some inferences or make
    generalization, and
ď‚—   5. To estimate parameters.
   Analysis means “computation of certain
    measures along with searching for pattern of
    relationships that exist among data groups”

ď‚—   Depending on the measurement and sampling
    procedures, the analysis of data are of two
    types.


ď‚— Statistical analysis (inferential)
ď‚— Non-statistical (Descriptive)
ď‚—   Descriptive analysis is largely the study of
    distributions of one variable.

ď‚—   This study provides us with profiles of
    companies, work groups, persons and other
    subjects on any of a multiple of characteristics
    such as size. Composition, efficiency,
    preferences, etc.”

ď‚—   Statistical analysis is always more precise and
    objective.

ď‚—   Selection and choice of statistical tool depends
    on three factors.
ď‚—   1) DEPENDING ON TYPE OF MEASUREMENT
      a) Nominal measurement
      b) Ordinal measurement
      c) Interval measurement
      d) Ratio measurement

For nominal and ordinary measurement, we
  commonly use nonparametric tests.

For interval and ratio measurement, we commonly
  use parametric tests. Parametric statistics are
  commonly used tests.
ď‚—   For application of parametric test the data should
    fulfil two conditions:

    A) Homogeneity of variance
    B) variables involved must be true numerical

   Nonparametric tests are also called “distribution-
    free tests”. so these can be used in statistics
    involved in nominal measurements and ordinal
    measurements.
ď‚—   2) DEPENDING ON NUMBER OF VARIABLES TO
    BE ANALYZED:-

    a) One variable - unidimensional analysis
    b) Two variable – bivariate analysis
    c) More than two variables – multivariate
                                  analysis

If investigator is interested in describing a single
   population, he should use “unidimensional analysis”.

If investigator wants to study interrelationship between
   two variables he should use “bivariate analysis”.
ď‚—   If the investigator wants to study
    interrelationship between more than two
    variables, he should use Regression analysis and
    multiple discriminant analysis.

ď‚—   It should be remembered that regression
    technique is useful only if both variables
    (independent variable and dependent variable)
    are interval variables.
ď‚—   3) DEPENDING ON TYPE OF ANALYSIS TO
    BE DONE:-

ď‚—   a) Requires estimating a parameter
      1. Point estimate (measures of
          central tendency)
      2. Interval estimate (measures of dispersion)

ď‚—   b) Testing of hypothesis
       1.„t‟ test.
       2. ANOVA test
       3. chi-square test
ď‚—   Choice of statistical technique depends on the
    type of “statistical inference” the researcher
    wants.

ď‚—   Point estimate:-
    A single statistic is used as an estimate of a
    parameter ( mean, median, mode )

ď‚—   Interval estimate:-
    An interval within which the true value of a
    parameter of a population is stated to lie with a
    predetermined probability on the basis of
    sampling statistics. (also called range)
ď‚—    STATISTICAL ANALYSIS:-

 It includes various ”tests of significance” and
  “testing of hypothesis”
ď‚— It is also useful in the estimation of population
  values.
ď‚— Ex.
  Z test, „t‟ test, chi-square test etc.

Now there are two type of analysis.

1.   Correlation analysis
2.   Causal analysis
ď‚—   Correlation analysis:-
    It is useful to study the correlation between two or
    more variables and determining the amount of
    correlation between two or more variables.

ď‚—   It is relatively more important in most social and
    business researches.

ď‚—   Causal analysis:-
    It is concerned with the study of how one or more
    variables(independent variable) affect the changes in
    another variable(dependent variable). This analysis
    can be termed as “Regression analysis”

ď‚—   It is considered relatively more important in
    experimental researches.
ď‚—   MULTIVARIATE ANALYSIS:-

ď‚—   A) Multiple regression analysis:
     In this type of analysis we measure the changes
    in dependent variable, with changes in two or
    more than two independent variables.

ď‚—   This is done by finding a constant called
    “regression coefficient”.

ď‚—   So the main objective of multiple regression
    analysis is to make a prediction about the
    dependent variable based on its “covariance”
    with two or more independent variables.
ď‚—   B) Multiple discriminant analysis:

ď‚—   This analysis is used when there is a single
    dependent variable, which needs to be classified
    into two or more than two groups.

ď‚—   This analysis is used when we want to predict an
    entity‟s possibility of belonging to a particular
    group based on several predictor variables.

ď‚—   C) Multivariate analysis of variance
       (Multi ANOVA):

    ANOVA is the ratio of “variance among the
    groups” to “variance within the groups”
   Multi ANOVA is an extension of “ two-way
    ANOVA”, wherein the ratio of among group
    variance to within group variance is worked out
    on set of variables.
ď‚—   When we want to use ANOVA or Multi ANOVA,
    the data should meet three assumptions.

ď‚—   These are:

ď‚—   1) Selection of subjects on the basis of random
    sampling.
ď‚—   2) Existence of homogeneity of variance between
    groups.
ď‚—   3) Variables under study should follow normal
    distribution.
ď‚—   D) Canonical analysis:

ď‚—   This analysis is useful in case of both measurable
    and non-measurable variables.

ď‚—   for the purpose of simultaneously predicting a
    set of dependent variables from their joint
    covariance with a set of independent variables.
Analysis of data (pratik)

Analysis of data (pratik)

  • 1.
  • 2.
    ď‚— The analysis and interpretation of data represent the application of deductive and inductive logic to the research process. ď‚— The data are often classified by division into, subgroups, and are then analyzed and synthesized in such a way that hypothesis may be verified or rejected. ď‚— The final result may be a new principle or generalization. Data are examined in terms of comparison between the more homogeneous segments within the group any by comparison with some outside criteria.
  • 3.
     NEED FOR ANALYSIS OF DATA OR TREATMENT OF DATA  After administering and scoring research tools scripts, data collected and organized. The collected data are known as „raw data.‟  The raw data are meaningless unless certain statistical treatment is given to them.  Analysis of data means to make the raw data meaningful or to draw some results from the data after the proper treatment.
  • 4.
    ď‚— Thus, the analysis of data serves the following main functions: ď‚— 1. To make the raw data meaningful, ď‚— 2. To test null hypothesis, ď‚— 3. To obtain the significant results, ď‚— 4. To draw some inferences or make generalization, and ď‚— 5. To estimate parameters.
  • 5.
     Analysis means “computation of certain measures along with searching for pattern of relationships that exist among data groups”  Depending on the measurement and sampling procedures, the analysis of data are of two types.  Statistical analysis (inferential)  Non-statistical (Descriptive)
  • 6.
     Descriptive analysis is largely the study of distributions of one variable.  This study provides us with profiles of companies, work groups, persons and other subjects on any of a multiple of characteristics such as size. Composition, efficiency, preferences, etc.”  Statistical analysis is always more precise and objective.  Selection and choice of statistical tool depends on three factors.
  • 7.
    ď‚— 1) DEPENDING ON TYPE OF MEASUREMENT a) Nominal measurement b) Ordinal measurement c) Interval measurement d) Ratio measurement For nominal and ordinary measurement, we commonly use nonparametric tests. For interval and ratio measurement, we commonly use parametric tests. Parametric statistics are commonly used tests.
  • 8.
     For application of parametric test the data should fulfil two conditions: A) Homogeneity of variance B) variables involved must be true numerical  Nonparametric tests are also called “distribution- free tests”. so these can be used in statistics involved in nominal measurements and ordinal measurements.
  • 9.
     2) DEPENDING ON NUMBER OF VARIABLES TO BE ANALYZED:- a) One variable - unidimensional analysis b) Two variable – bivariate analysis c) More than two variables – multivariate analysis If investigator is interested in describing a single population, he should use “unidimensional analysis”. If investigator wants to study interrelationship between two variables he should use “bivariate analysis”.
  • 10.
    ď‚— If the investigator wants to study interrelationship between more than two variables, he should use Regression analysis and multiple discriminant analysis. ď‚— It should be remembered that regression technique is useful only if both variables (independent variable and dependent variable) are interval variables.
  • 11.
     3) DEPENDING ON TYPE OF ANALYSIS TO BE DONE:-  a) Requires estimating a parameter 1. Point estimate (measures of central tendency) 2. Interval estimate (measures of dispersion)  b) Testing of hypothesis 1.„t‟ test. 2. ANOVA test 3. chi-square test
  • 12.
     Choice of statistical technique depends on the type of “statistical inference” the researcher wants.  Point estimate:- A single statistic is used as an estimate of a parameter ( mean, median, mode )  Interval estimate:- An interval within which the true value of a parameter of a population is stated to lie with a predetermined probability on the basis of sampling statistics. (also called range)
  • 13.
     STATISTICAL ANALYSIS:-  It includes various ”tests of significance” and “testing of hypothesis”  It is also useful in the estimation of population values.  Ex. Z test, „t‟ test, chi-square test etc. Now there are two type of analysis. 1. Correlation analysis 2. Causal analysis
  • 14.
     Correlation analysis:- It is useful to study the correlation between two or more variables and determining the amount of correlation between two or more variables.  It is relatively more important in most social and business researches.  Causal analysis:- It is concerned with the study of how one or more variables(independent variable) affect the changes in another variable(dependent variable). This analysis can be termed as “Regression analysis”  It is considered relatively more important in experimental researches.
  • 15.
     MULTIVARIATE ANALYSIS:-  A) Multiple regression analysis: In this type of analysis we measure the changes in dependent variable, with changes in two or more than two independent variables.  This is done by finding a constant called “regression coefficient”.  So the main objective of multiple regression analysis is to make a prediction about the dependent variable based on its “covariance” with two or more independent variables.
  • 16.
     B) Multiple discriminant analysis:  This analysis is used when there is a single dependent variable, which needs to be classified into two or more than two groups.  This analysis is used when we want to predict an entity‟s possibility of belonging to a particular group based on several predictor variables.  C) Multivariate analysis of variance (Multi ANOVA): ANOVA is the ratio of “variance among the groups” to “variance within the groups”
  • 17.
     Multi ANOVA is an extension of “ two-way ANOVA”, wherein the ratio of among group variance to within group variance is worked out on set of variables.  When we want to use ANOVA or Multi ANOVA, the data should meet three assumptions.  These are:  1) Selection of subjects on the basis of random sampling.  2) Existence of homogeneity of variance between groups.  3) Variables under study should follow normal distribution.
  • 18.
    ď‚— D) Canonical analysis: ď‚— This analysis is useful in case of both measurable and non-measurable variables. ď‚— for the purpose of simultaneously predicting a set of dependent variables from their joint covariance with a set of independent variables.