co-relational research
TRANSCRIPT
7/30/2019 co-relational research
http://slidepdf.com/reader/full/co-relational-research 1/34
Educational Research
Chapter 7Correlational Research
Gay, Mills, and Airasian
7/30/2019 co-relational research
http://slidepdf.com/reader/full/co-relational-research 2/34
Topics to Be Discussed Definition, purpose, and limitation of
correlational research
Correlation coefficients and theirsignificance
Process of conducting correlational
research Relationship studies
Prediction studies
7/30/2019 co-relational research
http://slidepdf.com/reader/full/co-relational-research 3/34
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
7/30/2019 co-relational research
http://slidepdf.com/reader/full/co-relational-research 4/34
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 Inefficiency and difficulty interpreting the
results from a shotgun approach
Objective 2.1
7/30/2019 co-relational research
http://slidepdf.com/reader/full/co-relational-research 5/34
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
7/30/2019 co-relational research
http://slidepdf.com/reader/full/co-relational-research 6/34
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
7/30/2019 co-relational research
http://slidepdf.com/reader/full/co-relational-research 7/34
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
7/30/2019 co-relational research
http://slidepdf.com/reader/full/co-relational-research 8/34
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
7/30/2019 co-relational research
http://slidepdf.com/reader/full/co-relational-research 9/34
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
7/30/2019 co-relational research
http://slidepdf.com/reader/full/co-relational-research 10/34
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 thestrongest, the first or last?
Objective 3.3 & 3.5
7/30/2019 co-relational research
http://slidepdf.com/reader/full/co-relational-research 11/34
Correlation Coefficients Scatterplots
Graphical presentations of correlations
Example of predicting from an attitudescale – EX 1 – to an achievement test – EX 2
Predictor variable - EX1 - is on thehorizontal axis
Criterion variable - EX 2 - is on the verticalaxis
Objective 3.4
7/30/2019 co-relational research
http://slidepdf.com/reader/full/co-relational-research 12/34
An Example of a ScatterplotLinear Regression
30.00 40.00 50.00
ex 1
30.00
35.00
40.00
45.00
50.00
e x 2
ex2 = 11.23 + 0.72 * ex1R-Square = 0.66
Objective 3.4
7/30/2019 co-relational research
http://slidepdf.com/reader/full/co-relational-research 13/34
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 - r 2
Examples
If r = .50 then r 2 = .25
25% of the variance in the criterion can be explainedby the predictor
If r = .70 then r 2 = .49
49% of the variance in the criterion can be explainedby the predictor
Objectives 3.6 & 3.7
7/30/2019 co-relational research
http://slidepdf.com/reader/full/co-relational-research 14/34
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
7/30/2019 co-relational research
http://slidepdf.com/reader/full/co-relational-research 15/34
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 andpredicts 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 predictsonly .09 of the variation in the criterion scores
Objectives 4.2 & 4.4
7/30/2019 co-relational research
http://slidepdf.com/reader/full/co-relational-research 16/34
Relationship Studies General purpose
Gain insight into variables that are related to othervariables 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 dependentvariable in experimental studies
Objectives 5.1 & 5.2
7/30/2019 co-relational research
http://slidepdf.com/reader/full/co-relational-research 17/34
Conducting Relationship Studies Identify a set of variables
Limit to those variables logically related to the criterion
Avoid the shotgun approach
Possibility of erroneous relationships
Issues related to determining statistical significance
Identify a population and select a sample
Identify appropriate instruments for measuring eachvariable
Collect data for each instrument from each subject
Compute the appropriate correlation coefficient
Objective 6.1
7/30/2019 co-relational research
http://slidepdf.com/reader/full/co-relational-research 18/34
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 criterionvariables
Rank in class and rank on a final exam
Phi coefficient – dichotomous predictor and criterionvariables
Gender and pass/fail status on a high stakes test
See Table 7.2
Objectives 7.1, 7.2, & 7.3
7/30/2019 co-relational research
http://slidepdf.com/reader/full/co-relational-research 19/34
Linear and Curvilinear Relationships Linear relationships
Plots of the scores on two variables are bestdescribed 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
7/30/2019 co-relational research
http://slidepdf.com/reader/full/co-relational-research 20/34
An Example of a Linear Relationship
Linear Regression
30.00 40.00 50.00
ex 1
0.7000
0.8000
0.9000
1.0000
f p
fp = 0.39 + 0.01 * ex1R-Square = 0.80
Objective 8.4
7/30/2019 co-relational research
http://slidepdf.com/reader/full/co-relational-research 21/34
An Example of a Curvilinear Relationship
LLR Smoother
2.00 4.00 6.00 8.00 10.00
study
0.00
25.00
50.00
75.00
100.00
s c o r e
Objective 8.4
7/30/2019 co-relational research
http://slidepdf.com/reader/full/co-relational-research 22/34
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 theyare being obscured by the analysis of the data for thetotal sample
Reduces the size of the sample
Potentially reduces variation in the scores
Objective 9.1
7/30/2019 co-relational research
http://slidepdf.com/reader/full/co-relational-research 23/34
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
7/30/2019 co-relational research
http://slidepdf.com/reader/full/co-relational-research 24/34
Prediction Studies Attempts to describe the predictive
relationships between or amongvariables
The predictor variable is the variable fromwhich the researcher is predicting
The criterion variable is the variable towhich the researcher is predicting
Objectives 10.1 & 10.2
7/30/2019 co-relational research
http://slidepdf.com/reader/full/co-relational-research 25/34
Prediction Studies Three purposes
Facilitates decisions about individuals tohelp a selection decision
Tests variables believed to be goodpredictors of a criterion
Determines the predictive validity of aninstrument
Objective 11.1
7/30/2019 co-relational research
http://slidepdf.com/reader/full/co-relational-research 26/34
Prediction Studies Single and multiple predictors
Linear regression - one predictor and onecriterion Y’ = a + b X
r 2
Multiple regression – more than onepredictor and one criterion Y’ = a + b X1 + b X2 + … + b Xi
r 2 or the coefficient of determination
Objective 11.4
7/30/2019 co-relational research
http://slidepdf.com/reader/full/co-relational-research 27/34
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 eachvariable 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)
7/30/2019 co-relational research
http://slidepdf.com/reader/full/co-relational-research 28/34
Conducting a Prediction Study Issues of concern
Shrinkage – the tendency of a prediction
equation to become less accurate whenused with a group other than the one onwhich the equation was originallydeveloped
Cross validation – validation of a predictionequation with another group of subjects toidentify problematic variables
Objective 11.3
7/30/2019 co-relational research
http://slidepdf.com/reader/full/co-relational-research 29/34
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 betweencollecting the predictor and criterion variables
Criterion variables defined in general terms (e.g.,teacher effectiveness, success in school) tend tohave lower prediction accuracy than those definedvery narrowly (e.g., overall GPA, test scores)
Objective 11.5
7/30/2019 co-relational research
http://slidepdf.com/reader/full/co-relational-research 30/34
Differences between Types of Studies Correlational research is a general category
that is usually discussed in terms of twovariables
Relationship studies develop insight into therelationships between several variables The measurement of all variables occurs at about
the same time
Predictive studies involve the predictiverelationships between or among variables The predictor variables are collected long before
the criterion variableObjectives 11.2 & 11.3
7/30/2019 co-relational research
http://slidepdf.com/reader/full/co-relational-research 31/34
Other Correlation Analyses Path analysis
Investigates the patterns of relationships among anumber of variables
Results in a diagram that indicates the specificmanner 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
7/30/2019 co-relational research
http://slidepdf.com/reader/full/co-relational-research 32/34
Other Correlation Analyses Discriminant function analysis
Similar to multiple regression except thatthe criterion variable is categorical
Typically used to predict groupmembership
High or low anxiety
Achievers or non-achievers
Objective 13.2
7/30/2019 co-relational research
http://slidepdf.com/reader/full/co-relational-research 33/34
Other Correlation Analyses Cannonical correlation
An extension of multiple regression in which morethan one predictor variable and more than onecriterion variable are used
Factor analysis
A correlational analysis used to take a largenumber of variables and group them into a smallernumber of clusters of similar variables calledfactors
Objectives 13.3 & 13.4
7/30/2019 co-relational research
http://slidepdf.com/reader/full/co-relational-research 34/34
A Checklist of Questions Was the correct correlation coefficient
used?
Is the validity and reliability of theinstruments acceptable?
Is there a restricted range of scores?
How large is the sample?