Statistics in Perspective

Chapter Twelve
Dr Nek Kamal Yeop Yunus
Faculty of business & economics
Sultan Idris Education University

McGraw-Hill

© 2006 The McGraw-Hill Companies, Inc. All rights

Statistics in Perspective Chapter Twelve McGraw-Hill © 2006 The McGraw-Hill Companies. Inc. All rights .

Inc.Approaches to Research Educational Research is performed in one of two ways:  1) 2)   Two or more groups are compared Variables within one group are related The data in a study may be either quantitative or categorical. All rights . Four different combinations of research are possible.1) All groups are made up of individual units McGraw-Hill © 2006 The McGraw-Hill Companies. (see Figure 12.

1) McGraw-Hill © 2006 The McGraw-Hill Companies. All rights . Inc.Combinations of Data and Approaches (Figure 12.

2) McGraw-Hill © 2006 The McGraw-Hill Companies. Inc. All rights .A Difference that Doesn’t Make a Difference (Figure 12.

researchers can compare them through the following:     Frequency polygons Calculation of averages Calculations of spreads Recommendations are made to perform the following:   McGraw-Hill Prepare a frequency polygon of each group’s score Use these polygons to decide which measure of central tendency is appropriate to calculate © 2006 The McGraw-Hill Companies.Comparing Groups: Quantitative Data  When comparing two or more groups using quantitative data. All rights . Inc.

Frequency Polygons McGraw-Hill (Figure 12. All rights . Inc.3) © 2006 The McGraw-Hill Companies.

90% Confidence Interval McGraw-Hill (Figure 12. All rights . Inc.4) © 2006 The McGraw-Hill Companies.

Considerations of the following can assist with interpreting data more clearly:    McGraw-Hill Information based on known groups Calculate the effect size of the groups Use inferential statistics e. Inc.g. t-tests. ANOVA. All rights . interpretations must follow. etc. © 2006 The McGraw-Hill Companies.)   Once the descriptive statistics have been calculated.Comparing Groups: Quantitative Data (cont.

00 to . All rights . .60 Large enough to be of practical as well as theoretical use. a McGraw-Hill © 2006 The McGraw-Hill Companies.81 to above Possibly an error in calculation.a . . even correlations this small may have a predictive value.40 Of little practical importance except in unusual circumstances.80 Very important.41 to .Interpretation of Correlation when Testing Research Hypotheses (Table 12. if not. perhaps of theoretical value. When selecting a very few people from a large group.1) Magnitude of r Interpretation . but rarely obtained in educational research. a very sizable relationship. Inc.61 to .

All rights . Inc.Scatterplots with a Pearson r of .5 (Figure 12.5) McGraw-Hill © 2006 The McGraw-Hill Companies.

McGraw-Hill © 2006 The McGraw-Hill Companies. groups may be compared through the following methods:      Percentages/Proportions Frequencies Crossbreak Tables A limitation of categorical data is that such evaluations are even harder than with quantitative data. The use of the summary statistic known as contingency coefficient is recommended but data must be presented in crossbreak tables. All rights .Comparing Groups: Categorical Data  When using categorical data. Inc.

precise values Report confidence intervals rather than significance levels if possible Use both parametric and non-parametric techniques to analyze data rather than either one alone © 2006 The McGraw-Hill Companies. Inc.Recap of Recommendations      Use graphic techniques before calculating numerical summary indices Use both graphs and summary indices to interpret results of a study Make use of external criteria to assess the magnitude of a relationship Use professional consensus when evaluating magnitude of effect size Consider using inferential statistics McGraw-Hill     Use tests of statistical significance only to evaluate generalizability When random sampling has not occurred. All rights . treat probabilities as approximations or crude indices vs.

Inc. All rights . Any questions? McGraw-Hill © 2006 The McGraw-Hill Companies.

All rights . Inc. Thank You McGraw-Hill © 2006 The McGraw-Hill Companies.