Download presentation

Presentation is loading. Please wait.

Published byJameson Baile Modified about 1 year ago

1
Copyright © Allyn & Bacon (2007) Correlational and Differential Research Graziano and Raulin Research Methods: Chapter 7 This multimedia product and its contents are protected under copyright law. The following are prohibited by law: (1) Any public performance or display, including transmission of any image over a network; (2) Preparation of any derivative work, including the extraction, in whole or in part, of any images; (3) Any rental, lease, or lending of the program.

2
Copyright © Allyn & Bacon (2007) Correlational Research Quantifies the strength of the relationship between two or more variables Quantifies the strength of the relationship between two or more variables Value of correlational research Value of correlational research –Correlations can be used for prediction –Evidence consistent or inconsistent with a theory Cannot prove a theory, but could negate a theory Cannot prove a theory, but could negate a theory Note: Correlations CANNOT establish causation Note: Correlations CANNOT establish causation

3
Copyright © Allyn & Bacon (2007) Correlation and Causation If A and B are correlated, then … If A and B are correlated, then … –A could cause B –B could cause A –Another variable could cause both A and B Must be cautious in drawing conclusions Must be cautious in drawing conclusions

4
Copyright © Allyn & Bacon (2007) Differential Research Methods Compare two or more preexisting groups Compare two or more preexisting groups Similar to both correlational and experimental research Similar to both correlational and experimental research –Same form as experimental research –Conceptually similar to correlational research (variables measured, but not manipulated) Cross-sectional design in developmental research is differential research Cross-sectional design in developmental research is differential research

5
Copyright © Allyn & Bacon (2007) Cross-Sectional and Longitudinal Research Cross-sectional designs are faster Cross-sectional designs are faster –Can test many age groups simultaneously But cohort effects can be a problem But cohort effects can be a problem –Defined as “shared life experiences of people of a given age that lead them to behave similarly to others their age and different from people of other ages” Longitudinal designs are essentially time- series designs Longitudinal designs are essentially time- series designs

6
Copyright © Allyn & Bacon (2007) Artifacts and Confounding Variables Confounding occurs when two variables vary together Confounding occurs when two variables vary together –Need to have them vary independently, usually by holding all but one variable constant –Failing to provide this control could result in artifactual findings –Procedures standardized for this reason Comparing groups is reasonable ONLY IF we standardized the measurement procedures Comparing groups is reasonable ONLY IF we standardized the measurement procedures

7
Copyright © Allyn & Bacon (2007) Correlational versus Differential Both involve the measurement, but not manipulation, of variables Both involve the measurement, but not manipulation, of variables –Therefore, neither is able to establish causation Differential is higher constraint because Differential is higher constraint because –The researcher can select the comparison group(s) to control at least some of the potential confounding variables, thus providing stronger evidence for a theory

8
Copyright © Allyn & Bacon (2007) When to Use Each Method Correlational Method Correlational Method –When we are interested in knowing the strength of a relationship for predictive purposes –Often included to help interpret the primary findings of a study Differential Research Differential Research –When the manipulation of an independent variable is impractical, impossible, or unethical –Then we rely on comparing preexisting groups

9
Copyright © Allyn & Bacon (2007) Doing Correlational Research Steps in conducting correlational research Steps in conducting correlational research –Developing the problem statement –Measuring the Variables –Obtaining the Sample –Analyzing the Data –Interpreting the Results Correlational research is often embedded in larger studies Correlational research is often embedded in larger studies

10
Copyright © Allyn & Bacon (2007) Developing a Problem Statement “What is the relationship between variable X and variable Y?” “What is the relationship between variable X and variable Y?” Often want to correlate available demographic variables with the dependent measures or intercorrelate the dependent measures in higher-constraint research Often want to correlate available demographic variables with the dependent measures or intercorrelate the dependent measures in higher-constraint research –Useful in detecting confounding variables –Provides hypotheses for later research

11
Copyright © Allyn & Bacon (2007) Measuring the Variables Need to use reliable and valid measures Need to use reliable and valid measures Need to control Need to control –Experimenter expectancy Researchers tending to see what they expect to see Researchers tending to see what they expect to see –Experimenter reactivity Researchers unconsciously influencing participants Researchers unconsciously influencing participants –Measurement reactivity Participants responding differently because they know they are being observed Participants responding differently because they know they are being observed

12
Copyright © Allyn & Bacon (2007) Controlling these effects Experimenter expectancy Experimenter expectancy –Use objective measures whenever possible Experimenter reactivity Experimenter reactivity –Minimize experimenter contact Measurement reactivity Measurement reactivity –Use filler items to distract participants –Use unobtrusive measures when possible –Separate the measurements in time

13
Copyright © Allyn & Bacon (2007) Sampling Considerations Want the sample to be representative Want the sample to be representative Is the observed relationship the same in each subpopulation? Is the observed relationship the same in each subpopulation? –If we suspect such differences, we should compute the correlation in each subpopulation –Moderator Variable: a variable that seems to modify the relationship between other variables e.g., sex: males and females showing different patterns of relationship between variables e.g., sex: males and females showing different patterns of relationship between variables

14
Copyright © Allyn & Bacon (2007) Analyzing the Data Correlations range from to Correlations range from to –Size indicates strength of the relationship –Sign indicates direction of the relationship Many types of correlations Many types of correlations –Pearson product-moment correlation –Spearman rank-order correlation –Advanced techniques (multiple correlation, canonical correlation, partial correlation, path analysis)

15
Copyright © Allyn & Bacon (2007) Interpreting the Data Note size and sign of correlation Note size and sign of correlation –Indicates strength and direction of relationship Is the correlation significantly different from zero (i.e., evidence for a relationship)? Is the correlation significantly different from zero (i.e., evidence for a relationship)? –Is the p value < alpha? Coefficient of Determination Coefficient of Determination –r 2 indicates the proportion of variance accounted for

16
Copyright © Allyn & Bacon (2007) Doing Differential Research Developing the problem statement Developing the problem statement Measuring the variables Measuring the variables Selecting appropriate control groups Selecting appropriate control groups Obtaining the sample Obtaining the sample Analyzing the data Analyzing the data Interpreting the results Interpreting the results

17
Copyright © Allyn & Bacon (2007) Developing a Problem Statement “Does Group A differ from Group B?” “Does Group A differ from Group B?” Developing good problem statements Developing good problem statements –Select theoretically interesting groups to compare –Compare them on theoretically interesting variables –Best to compare groups that differ on only a single variable if possible –Several comparisons are best

18
Copyright © Allyn & Bacon (2007) Measuring the Variables Dependent variable is usually continuous, but could be categorical Dependent variable is usually continuous, but could be categorical Independent variable is categorical or is a continuous variable converted to categories Independent variable is categorical or is a continuous variable converted to categories –Unlike experimental research, the independent variable is measured, rather than manipulated Need operational definitions for Need operational definitions for –Dependent Variable –Independent variables

19
Copyright © Allyn & Bacon (2007) Selecting Control Groups Select control groups to avoid confounding Select control groups to avoid confounding –A variable can confound results only if a) it affects the scores on the dependent variable b) the groups differ on this variable Ideal control group is identical to experimental group on all variables except the variable that defines the groups Ideal control group is identical to experimental group on all variables except the variable that defines the groups –Rarely possible, so multiple comparisons groups are typical

20
Copyright © Allyn & Bacon (2007) Sampling of Participants Like all research, we want representative sampling to permit generalization Like all research, we want representative sampling to permit generalization Many factors can bias sampling Many factors can bias sampling –Where we have access to participants –How we go about identifying participants –Even factors like time of day that we sample Participants who drop out of the study can limit generalizability Participants who drop out of the study can limit generalizability

21
Copyright © Allyn & Bacon (2007) Analyzing the Data Same procedures as those used to analyze experimental research Same procedures as those used to analyze experimental research Type of analysis depends on the number of groups and the level of measurement Type of analysis depends on the number of groups and the level of measurement –Score data: t-test or ANOVA –Ordinal data: Mann-Whitney U-test –Nominal data: Chi square

22
Copyright © Allyn & Bacon (2007) Interpreting the Results Reject null hypothesis (of no group difference) if the p < alpha Reject null hypothesis (of no group difference) if the p < alpha Difficult to draw a strong conclusion from differential research Difficult to draw a strong conclusion from differential research –Sampling considerations –Unlikely that all potential confounding variables will have been adequately controlled

23
Copyright © Allyn & Bacon (2007) Blanchard et al. (2001)

24
Copyright © Allyn & Bacon (2007) Limitations of these Methods Problems in determining causation Problems in determining causation –A correlation does not imply causality If A and B are correlated, then If A and B are correlated, then –A could cause B –B could cause A –Some other variable could cause both Confounding variables Confounding variables –Without experimental control, it is virtually impossible to avoid confounding variables

25
Copyright © Allyn & Bacon (2007) Summary Both correlational and differential research involve measuring the relationship between variables Both correlational and differential research involve measuring the relationship between variables Drawing causal inferences is risky Drawing causal inferences is risky Selecting appropriate control groups in the differential research design can control some, but typically not all, potential confounding variables Selecting appropriate control groups in the differential research design can control some, but typically not all, potential confounding variables

Similar presentations

© 2016 SlidePlayer.com Inc.

All rights reserved.

Ads by Google