You are on page 1of 34

# Educational Research

Chapter 7
Correlational Research

## Gay, Mills, and Airasian

Topics to Be Discussed
 Definition, purpose, and limitation of
correlational research
 Correlation coefficients and their
significance
 Process of conducting correlational
research
 Relationship studies
 Prediction studies
Correlational Research
 Definition
 Whether and to what degree variables are
related
 Purpose
 Determine relationships
 Make predictions
 Limitation
 Cannot indicate cause and effect

## Objectives 1.1, 1.2, & 1.3

The Process
 Problem selection
 Variables to be correlated are selected on the
basis of some rationale
 Math attitudes and math achievement
 Teachers’ sense of efficacy and their effectiveness
 Increases the ability to meaningfully interpret
results

Objective 2.1
The Process
 Participant and instrument selection
 Minimum of 30 subjects
 Instruments must be valid and reliable
 Higher validity and reliability requires smaller samples
 Lower validity and reliability requires larger samples
 Design and procedures
 Collect data on two or more variables for each
subject
 Data analysis
 Compute the appropriate correlation coefficient
Objectives 2.2 & 2.3
Correlation Coefficients
 A correlation coefficient identifies the
size and direction of a relationship
 Size/magnitude
 Ranges from 0.00 – 1.00
 Direction
 Positive or negative

## Objectives 3.1, 3.2, & 3.3

Correlation Coefficients
 Interpreting the size of correlations
 General rule
 Less than .35 is a low correlation
 Between .36 and .65 is a moderate correlation
 Above .66 is a high correlation
 Predictions
 Between .60 and .70 are adequate for group
predictions
 Above .80 is adequate for individual predictions
Objective 3.5
Correlation Coefficients
 Interpreting the size of correlations (cont.)
 Criterion-related validity
 Above .60 for affective scales is adequate
 Above .80 for tests is minimally acceptable
 Inter-rater reliability
 Above .90 is very good
 Between .80 and .89 is acceptable
 Between .70 and .79 is minimally acceptable
 Lower than .69 is problematic

Objective 3.5
Correlation Coefficients
 Interpreting the direction of correlations
 Direction
 Positive
 High scores on the predictor are associated with high
scores on the criterion
 Low scores on the predictor are associated with low
scores on the criterion
 Negative
 High scores on the predictor are associated with low
scores on the criterion
 Low scores on the predictor are associated with high
scores on the criterion
 Positive or negative does not mean good or bad

Objective 3.3
Correlation Coefficients
 Interpreting the size and direction of
correlations using the general rule
 +.95 is a strong positive correlation
 +.50 is a moderate positive correlation
 +.20 is a low positive correlation
 -.26 is a low negative correlation
 -.49 is a moderate negative correlation
 -.95 is a strong negative correlation
 Which of the correlations above is the
strongest, the first or last?
Objective 3.3 & 3.5
Correlation Coefficients
 Scatterplots
 Graphical presentations of correlations
 Example of predicting from an attitude
scale – EX 1 – to an achievement test –
EX 2
 Predictor variable - EX1 - is on the
horizontal axis
 Criterion variable - EX 2 - is on the vertical
axis
Objective 3.4
An Example of a Scatterplot
 
50.00 Linear Regression
ex2 = 11.23 + 0.72 * ex1 
R-Square = 0.66   
  
    

45.00
  

   
 
40.00 
   
ex2

35.00  
 

30.00

 

## 30.00 40.00 50.00

Objective 3.4
ex1
Correlation Coefficients
 Common variance
 Definition
 The extent to which variables vary in a systematic manner
 Interpreted as the percentage of variance in the criterion
variable explained by the predictor variable
 Computation
 The squared correlation coefficient - r2
 Examples
 If r = .50 then r = .25
2

##  25% of the variance in the criterion can be explained

by the predictor
 If r = .70 then r = .49
2

by the predictor

## Objectives 3.6 & 3.7

Statistical Significance
 Statistical significance
 Is the observed coefficient different from 0.00?
 Does the correlation represent a true relationship?
 Is the correlation only the result of chance?
 Determining statistical significance
 Consult a table of the critical values of r
 See Table A.2 in Appendix A
 Three common levels of significance
 .01 (1 chance out of 100)
 .05 (5 chances out of 100)
 .10 (10 chances out of 100)

## Objectives 4.1 & 4.3

Statistical Significance
 Sample size and statistical significance
 Small samples require higher correlations for significance
 Large samples require lower correlations for significance
 Practical significance and statistical significance
 Small correlation coefficients can be statistically significant even
though they have little practical significance
 +.20

##  Statistically significant at the .05 level if the sample is about 100

 Little or no practical significance because it is very low and
predicts only .04 of the variation in the criterion scores
 -.30
 Statistically significant at the .05 level if the sample is about 40
 Little or no practical significance because it is low and predicts
only .09 of the variation in the criterion scores

## Objectives 4.2 & 4.4

Relationship Studies
 General purpose
 Gain insight into variables that are related to other
variables relevant to educators
 Achievement
 Self-esteem
 Self-concept
 Two specific purposes
 Suggest subsequent interest in establishing cause
and effect between variables found to be related
 Control for variables related to the dependent
variable in experimental studies
Objectives 5.1 & 5.2
Conducting Relationship Studies
 Identify a set of variables
 Limit to those variables logically related to the criterion
 Identify a population and select a sample
 Identify appropriate instruments for measuring each
variable
 Collect data for each instrument from each subject
 Compute the appropriate correlation coefficient

Objective 6.1
Types of Correlation Coefficients
 The type of correlation coefficient depends on the
measurement level of the variables
 Pearson r - continuous predictor and criterion variables
 Math attitude and math achievement
 Spearman rho – ranked or ordinal predictor and criterion
variables
 Rank in class and rank on a final exam
 Phi coefficient – dichotomous predictor and criterion
variables
 Gender and pass/fail status on a high stakes test
 See Table 7.2

## Objectives 7.1, 7.2, & 7.3

Linear and Curvilinear Relationships
 Linear relationships
 Plots of the scores on two variables are best
described by a straight line
 Math scores and science scores
 Teacher efficacy and teacher effectiveness
 Curvilinear relationships
 Plots of scores on two variables are best described
by functions
 Age and athletic ability
 Anxiety and achievement
 Estimated by the eta correlation
Objectives 8.1, 8.2, & 8.3
An Example of a Linear Relationship

1.0000 
  Linear Regression
fp = 0.39 + 0.01 * ex1
  
R-Square = 0.80  

 
 

 
  
 
0.9000 
 

fp

0.8000 

 

0.7000 

## 30.00 40.00 50.00

ex1

Objective 8.4
An Example of a Curvilinear Relationship

100.00 LLR Smoother

75.00

score

50.00 

25.00 

0.00

## 2.00 4.00 6.00 8.00 10.00

study

Objective 8.4
Factors that Influence Correlations
 Sample size
 The larger the sample the higher the likelihood of
a high correlation
 Analysis of subgroups
 If the total sample consists of males and females each
gender represents a subgroup
 Results across subgroups can be different because they
are being obscured by the analysis of the data for the
total sample
 Reduces the size of the sample
 Potentially reduces variation in the scores

Objective 9.1
Factors that Influence Correlations
 Variation
 The greater the variation in scores the
higher the likelihood of a strong correlation
 The lower the variation in scores the
higher the likelihood of a weak correlation
 Attenuation
 Correlation coefficients are lower when the
instruments being used have low reliability
 A correction for attenuation is available
Objectives 9.2 & 9.3
Prediction Studies

##  Attempts to describe the predictive

relationships between or among
variables
 The predictor variable is the variable from
which the researcher is predicting
 The criterion variable is the variable to
which the researcher is predicting
Objectives 10.1 & 10.2
Prediction Studies

 Three purposes
 Facilitates decisions about individuals to
help a selection decision
 Tests variables believed to be good
predictors of a criterion
 Determines the predictive validity of an
instrument
Objective 11.1
Prediction Studies
 Single and multiple predictors
 Linear regression - one predictor and one
criterion
 Y’ = a + bX
 r2
 Multiple regression – more than one
predictor and one criterion
 Y’ = a + bX1 + bX2 + … + bXi
 r2 or the coefficient of determination
Objective 11.4
Conducting a Prediction Study
 Identify a set of variables
 Limit to those variables logically related to the criterion
 Identify a population and select a sample
 Identify appropriate instruments for measuring each
variable
 Ensure appropriate levels of validity and reliability
 Collect data for each instrument from each subject
 Typically data is collected at different points in time
 Compute the results
 The multiple regression coefficient

##  The multiple regression equation (i.e., the

prediction equation)
Conducting a Prediction Study
 Issues of concern
 Shrinkage – the tendency of a prediction
equation to become less accurate when
used with a group other than the one on
which the equation was originally
developed
 Cross validation – validation of a prediction
equation with another group of subjects to
identify problematic variables
Objective 11.3
Conducting a Prediction Study
 Issues of concern (cont.)
 Errors of measurement (e.g., low validity or
reliability) diminish the accuracy of the prediction
 Intervening variables can influence the predictive
process if there is too much time between
collecting the predictor and criterion variables
 Criterion variables defined in general terms (e.g.,
teacher effectiveness, success in school) tend to
have lower prediction accuracy than those defined
very narrowly (e.g., overall GPA, test scores)
Objective 11.5
Differences between Types of Studies
 Correlational research is a general category
that is usually discussed in terms of two
variables
 Relationship studies develop insight into the
relationships between several variables
 The measurement of all variables occurs at about
the same time
 Predictive studies involve the predictive
relationships between or among variables
 The predictor variables are collected long before
the criterion variable
Objectives 11.2 & 11.3
Other Correlation Analyses
 Path analysis
 Investigates the patterns of relationships among a
number of variables
 Results in a diagram that indicates the specific
manner by which variables are related (i.e., paths)
and the strength of those relationships
 An extension of this analysis is structural equation
modeling (SEM)
 Clarifies the direct and indirect relationships among
variables based on underlying theoretical constructs
 More precise than path analysis
 Often known as LISREL for the first computer program
used to conduct this analysis
Objective 13.1
Other Correlation Analyses

##  Discriminant function analysis

 Similar to multiple regression except that
the criterion variable is categorical
 Typically used to predict group
membership
 High or low anxiety
 Achievers or non-achievers

Objective 13.2
Other Correlation Analyses
 Cannonical correlation
 An extension of multiple regression in which more
than one predictor variable and more than one
criterion variable are used
 Factor analysis
 A correlational analysis used to take a large
number of variables and group them into a smaller
number of clusters of similar variables called
factors
Objectives 13.3 & 13.4
A Checklist of Questions
 Was the correct correlation coefficient
used?
 Is the validity and reliability of the
instruments acceptable?
 Is there a restricted range of scores?
 How large is the sample?