Reading...
Play button
Play button
Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
53 Cards in this Set
- Front
- Back
|
A study designed to establish the relationship between two variables is best described as
A. descriptive. B. correlational. C. experimental. D. single subject. |
B. Correlational.
This is the unique purpose of correlational research. No other design investigates non-causal relationships. |
|
|
Which of the following is NOT a purpose of correlational research?
A. To determine if a relation exists between two variables B. To determine the degree to which a relationship exists between two variables C. To make predictions D. To establish causal relationships |
D. To establish causal relationships
The major limitation to correlational research is the inability to control extraneous variables so cause and effect relationships can be established. |
|
|
Which of the following is NOT a reasonable approach to developing a correlational research problem?
A. Correlating two variables for which you've found some relationship discussed in the literature. B. Correlating 12 variables that make up a data set a colleague gave you. C. Attempting to understand the extent to which two variables discussed in the literature are related. D. Replicating the relationship found between two variables in a different population. |
B. Correlating 12 variables that make up a data set a colleague gave you.
A "shotgun" approach to correlational research is inefficient and statistically problematic. There must be a rationale for wanting to look at specific variables. |
|
|
What is the minimal sample size for a correlational study?
A. 15 B. 30 C. 50 D. 100 |
B. 30
This is the general rule for a minimum sample size in correlational research. |
|
|
What is the appropriate response to compensate for using a test with low validity and reliability as a criterion variable?
A. Decrease the number of subjects B. Apply the Spearman Brown formula to correct the correlations C. Disregard the validity and reliability issues D. Increase the number of subjects |
D. Increase the number of subjects
Low validity or reliability affects the correlation coefficient negatively. The researcher could have made a better choice for an instrument, but at this point increasing the sample size is the only alternative. |
|
|
Ms. Perkins calculated a correlation coefficient of .75. Which of the following reflects the best interpretation of this?
A. Weak negative B. Strong negative C. Weak positive D. Strong positive |
D. Strong positive
Generally correlations below .35 are considered weak, between .36 and .65 moderate, and above .75 strong. |
|
|
Smaller sample sizes than recommended can be used when
A. the researcher expects no relationship between variables. B. there are only two variables to control. C. the instruments have low validity and reliability. D. the instruments have high validity and reliability. |
D. the instruments have high validity and reliability.
High validity and reliability of the instruments used in correlational research are extremely important. |
|
|
In a study of the relationship between math anxiety and math aptitude, which of the following correlation coefficients is reasonable to expect?
A. +1.00 B. +.67 C. –.67 D. – 1.00 |
C. –.67
There is reason to suspect that higher levels of math anxiety are related to lower math achievement scores. |
|
|
If scores of assertiveness tend to be higher for students who have higher self-concept, which of the following is a reasonable correlational between assertiveness and self-concept?
A. +.21 B. +.78 C. –.78 D. –.21 |
B. +.78
High scores on one variable associated with high scores on a second variable describe a moderate to high positive correlation. |
|
|
Approximately what percentage of variation in the criterion variable is accounted for by the predictor variable if the correlation between them is .49?
A. 16% B. 24% C. 49% D. 70% |
B. 24%
The squared correlation coefficient (i.e., r2) is the percentage of common variance shared by the two variables. |
|
|
A scatterplot of a correlation of .10 would look most like a
A. straight line. B. ellipse. C. circle. D. square. |
C. circle.
The scatterplots for low correlations are circular; moderate and strong correlations have elliptical and straight line scatterplots. |
|
|
Which of the following correlation coefficients could be statistically significant?
A. –.45 B. .12 C. .34 D. All of these |
D. All of these
The statistical significance of a correlation coefficient is dependent on sample size. |
|
|
Using a sample size of 30 subjects, a researcher found a correlation of .30. This relationship was not statistically significant. If the researcher were to triple her sample size, which of the following is most likely?
A. The correlation would be statistically significant. B. The correlation would be larger. C. The correlation would be smaller. D. Nothing would change. |
B. The correlation would be statistically significant.
The statistical significance of correlations is directly connected to sample size. |
|
|
Which of the following is NOT a purpose of a relationship study?
A. Predict one variable from another B. Gain insight into variables that could be related to one another C. Suggest further research—particularly causal—into existing relationships D. Control for extraneous variables related to the dependent variable in an experimental study |
A. Predict one variable from another
The first alternative is the purpose of a predictive study. The last three alternatives reflect the general and specific purposes of correlational research. |
|
|
A serious problem when conducting a relationship study is the use of
A. theoretically related variables. B. logically connected variables. C. a "shotgun" approach to identifying variables. D. All of these |
C. a "shotgun" approach to identifying variables.
A "shotgun" approach to relationship studies increases the possibility of erroneous relationships and statistical significance. |
|
|
A Pearson Correlation is used under which condition?
A. The predictor and criterion variables are measured on a continuous scale. B. The predictor and criterion variables are measured on a dichotomous scale. C. The predictor and criterion variables are measured on a categorical scale. D. The predictor variable is ranked while the criterion variable is measured on a continuous scale. |
A. The predictor and criterion variables are measured on a continuous scale.
This is a definition of the Pearson Correlation. |
|
|
The Spearman rho is used under which condition?
A. The predictor and criterion variables are measured on a continuous scale. B. The predictor and criterion variables are measured on a dichotomous scale. C. The predictor and criterion variables are measured on a categorical scale. D. The predictor variable is ranked while the criterion variable is measured on a continuous scale. |
D. The predictor variable is ranked while the criterion variable is measured on a continuous scale.
This is the unique characteristic of the Spearman rho. |
|
|
Which of the following best describes the relationship between the popularity of a Top 20 song and the time since its release?
A. Linear B. Statistically significant C. Statistically non-significant D. Curvilinear |
D. Curvilinear
It is likely that the popularity of a song will drop drastically after a few months on the charts. |
|
|
A curvilinear relationship is measured by which correlation coefficient?
A. Pearson B. Spearman rho C. Eta D. Phi coefficient |
C. Eta
A curvilinear relationship cannot be measured accurately with any of the linear correlation coefficients. |
|
|
Which of the following correlations might produce different results for the total sample and subgroups within the sample?
A. Attitudes toward school for males and females B. Social studies scores for males and females C. Cardiovascular fitness levels for athletes and non-athletes D. Reading levels for students enrolled in extra-curricular activities and those not enrolled |
C. Cardiovascular fitness levels for athletes and non-athletes
There is a reason to expect athletes and non-athletes might differ statistically in cardiovascular fitness levels. |
|
|
The use of instruments with low reliabilities in relationship studies is known as.
A. low variability. B. attenuation. C. subgroup differences. D. all of these. |
B. attenuation.
This is the definition of attenuation. |
|
|
Low variability in scores will affect a correlation coefficient in which of the following ways?
A. Increases the correlation B. Changes the direction of the correlation C. Decreases the correlation D. None of the above |
C. Decreases the correlation
Lower levels of variability make it more difficult to differentiate "higher" students from "lower" students. This is likely to produce a lower correlation coefficient. |
|
|
The possible range of scores on a math achievement test was 0-100. The observed range for a sample of 100 students was 80-90. The possible range for the same students' science scores was 0-100, and the actual range was 45-49. The correlation between math and science scores was .29. The researcher would be most justified in concluding that the strength of the correlation was affected by
A. restriction of range. B. shrinkage. C. small sample size. D. attenuation. |
A. restriction of range.
This is a classic example of a restricted range of scores. |
|
|
In a prediction study of the relationship between age and cognitive development, the criterion variable would be
A. age after controlling for cognitive development. B. age. C. cognitive development. D. cognitive development after controlling for age. |
C. cognitive development.
The variable being used to predict is the predictor, and the variable being predicted is the criterion. |
|
|
Which of the following is a purpose of prediction studies?
A. It facilitates decisions about individuals to help a selection decision. B. It tests variables believed to be good predictors of a criterion. C. It determines the predictive validity of an instrument. D. All of these |
D. All of these
These are three legitimate purposes of a prediction study. |
|
|
Relationship and prediction studies differ with regard to
A. the direction of the correlation coefficients. B. the percentage of variation in the criterion variable explained by the predictor variable. C. the data collection of the criterion variables occurring at different times. D. the variability of scores. |
C. the data collection of the criterion variables occurring at different times.
In a relationship study the criterion variable is collected at the same time as the predictor variable. There is a relatively longer time between the collection of these in a prediction study. |
|
|
A researcher studied the predictors of dropout rates of urban school students over many years. He found his set of predictor variables explained 89% of the variability in the likelihood to quit school. In a study of 10,000 students in five rural school districts, the predictor variables accounted for only 45% of the variance. What is the most likely explanation for this finding?
A. Validity of instruments B. Attenuation C. Sample size D. Shrinkage |
D. Shrinkage
This is an example of the consequences of changing populations. |
|
|
When students' GPA in their freshmen year of college is predicted from their ACT scores and high school GPA, the analysis is known as
A. linear regression. B. correlation. C. r2 . D. multiple regression. |
D. multiple regression.
Predicting a criterion variable from more than one predictor variable is the purpose of multiple regression. |
|
|
Which of the following affects the accuracy of a prediction equation?
A. Errors of measurement B. Intervening variables C. Criterion variables measured in general terms D. All of these |
D. All of these
All of these can affect the prediction negatively. |
|
|
Which research method allows the researcher to reduce the number of variables studied?
A. Path analysis B. Discriminant function analysis C. Canonical correlation D. Factor analysis |
D. Factor analysis
Factor analysis uses correlational techniques to combine variables into common groups. |
|
|
A multiple regression predicting a categorical outcome is known as
A. path analysis. B. discriminant function analysis. C. canonical correlation. D. factor analysis. |
B. discriminant function analysis.
Discriminant function analysis uses multiple predictors to predict a categorical criterion. |
|
|
Correlating a set of four predictor variables with five criterion variables uses which analysis?
A. Path analysis B. Discriminant function analysis C. Canonical correlation D. Factor analysis |
C. Canonical correlation
This is an example of canonical correlation. |
|
|
Which correlational analysis is used to examine causal relationships?
A. Structural equation modeling B. Discriminant function analysis C. Canonical correlation D. Factor analysis |
A. Structural equation modeling
Structural equation modeling moves beyond path analysis to confirm causal relationships in a theoretical model. |
|
|
Type of research that involves collecting data to determine whether, and to what degree, a relationship exists between two or more quantifiable variables.
|
Correlational Research
|
|
|
The range of scores is the score variance or common variance?
|
Score Variance
|
|
|
The extent to which variables vary in a systematic manner is the score variance or the common variance?
|
The Common Variance:
The more common the variance, the higher the correlation coefficient. (closer to +/- 1.00) |
|
|
The probability that the results of a given relation would have occurred simply due to chance.
|
Statistical Significance
|
|
|
A measure of correlation used when both variables are expressed in terms of a true categorical dichotomy, such as gender, political affiliation, smoking status, or educational status. The two categories are labeled 1 and 0 or 1 and 2 (Numbering different categories, not different amounts!)
A. Pearson r B. Spearman rho C. Phi coefficient |
C. Phi Coefficient
|
|
|
A measure of correlation that is appropriate when the data for at least one variable are expressed as rank or ordinal data (Highest score is ranked 1, and then so on); better for smaller groups, cause you can take the time to rank them, and it’s easier to compute.
A. Pearson r B. Spearman rho C. Phi coefficient |
B. Spearman rho
|
|
|
A measure of correlation that is appropriate when both variables to be correlated are expressed as continuous (ratio or interval) data. (Use achievement or personality measures; Ex: IQ); better and more common for larger groups, cause you wouldn’t want to take the time to rank them all.
A. Pearson r B. Spearman rho C. Phi coefficient |
A. Pearson r
|
|
|
To avoid innaccurate estimates of relation in a correlational study, what must be kept?
|
Attenuation
|
|
|
The variable used to predict (high school grades or certification exam)
Predictor or Criterion |
Predictor
|
|
|
The variable that is predicted (college grades or principals’ evaluation)
Predictor or Criterion |
Criterion
|
|
|
This is the type of correlational study that attempts to determine which of a number of variables are most highly related to the criterion variable.
|
Prediciton Study
|
|
|
This is the type of correlational study that gains insight into variables that are related to a complex variable (like academic achievement, motivation, and self-concept).
|
Relationship Study
|
|
|
In a Prediction study, there is one predictor and one criterion:
o Y’ = a + bX o r2 What is the regression? |
Linear regression
|
|
|
In a Prediction study, there is more than one predictor and one criterion (makes for a more accurate predication):
o Y’ = a + bX1 + bX2 + … + bXi o r2 or the coefficient of determination What is the regression? |
Multiple regression
|
|
|
The tendency for the prediction to be less accurate for a group other than the one on which it was originally developed (prediction study may take a long period of time, so you might have participant loss; therefore, the initial finding may be the result of chance relations that will not be found again with another group of participants).
|
Shrinkage
|
|
|
You can do this to help validate shrinkage with a prediction equation with another group of subjects to identify problematic variables.
|
Cross-Validation
|
|
|
A variable where there is too much time between collecting the predictor and criterion variables
|
Intervening Variable
|
|
|
The result is a correlation coefficient, which is a decimal number ranging from -1.00 to +1.00. This represents the ________ and direction of the relation between variables.
|
size
|
|
|
Are coefficients of +/- .60 or +/-.70 adequate for group prediction?
|
Yes
|
|
|
Are coefficients of +/-.80 and higher adequate for individual prediction?
|
Yes
|
