Deck 13: Correlation and Regression: Are We Related

Question 1

Positive correlations are also described as indirect, whereas negative correlations are also described as direct.

Answer 1

False

Question 2

in a positive correlation, as the value of X increases, the value of Y also increases

Answer 2

True

Question 3

The strength of a correlation depends on whether it's positive or negative

Answer 3

False

Question 4

A correlation of .78 is stronger than a correlation of -.92.

Question 5

When testing a correlation, the null hypothesis states that there is no correlation

Question 6

A least squares regression line minimizes the distance from each datum in the dataset to it.

Question 7

In a regression equation, if b is positive, then the correlation between the predictor and outcome must be negatively correlated.

Question 8

In a regression equation, a stronger equation between the two variables would lead to more horizontal regression line.

Question 9

In a regression equation, if a is negative, then the correlation between the predictor and outcome must be negatively correlated.

Question 10

The coefficient of determination is used to explain the proportion of variance shared between two variables

Question 11

A strong negative correlation will lead to a large positive coefficient of determination

Question 12

Smaller coefficients of determination means that the two variables have a weaker relationship

Question 13

A correlation of .04 and -.04 will have the same coefficient of determination.

Question 14

Correlation might not mean causation, but if data is analyzed by an ANOVA or t-test, a significant result is a causal relationship.

Question 15

Correlations measures dependence between two variables

Question 16

If a correlation between two variables is -.2, the coefficient of determination is -.04

Question 17

When testing a correlation between 2 variables with 20 pairs of data, the degrees of freedom is 19

Question 18

In regression, the y-variable is often the predicted variable. The x-variable is used to predict the y-variable.

Question 19

The regression coefficient is negatively correlated with the correlation coefficient of the same two variables

Question 20

The Pearson correlation can be applied to linear relationships as well as curvilinear relationships

Question 21

A correlation coefficient is bounded between which two values?

A) 0.00 and 1.00
B) -1.00 and 1.00
C) There are no numeric boundaries. Correlations can go from weak to strong.
D) -100 and 100

Question 22

The correlation coefficient is a statistic that measures the degree of _________________ between two variables.

A) cause
B) association
C) validity
D) variability

Question 23

Which of the following is the strongest correlation?

A) -.77
B) .08
C) .36
D) .72

Question 24

Which of the following indicates the weakest correlation?

A) 0.11
B) -.28
C) -.97
D) .50

Question 25

A researcher is examining whether the amount of time newborns can sleep without being disturbed is related to parental stress. The researcher asked 30 parents to report the longest amount of time in number of minutes their newborn child has been able to sleep without waking up. Then the parents were asked to rate their stress levels.
How many degrees of freedom are there for this correlation?

A) 30
B) 29
C) 27
D) 28

Question 26

A researcher is examining whether the amount of time newborns can sleep without being disturbed is related to parental stress. The researcher asked 30 parents to report the longest amount of time in number of minutes their newborn child has been able to sleep without waking up. Then the parents were asked to rate their stress levels.
What is the critical value assuming α = .05?

A) .367
B) .361
C) .463
D) .471

Question 27

A researcher is examining whether the amount of time newborns can sleep without being disturbed is related to parental stress. The researcher asked 30 parents to report the longest amount of time in number of minutes their newborn child has been able to sleep without waking up. Then the parents were asked to rate their stress levels.
The researcher find a correlation of -.56. What does this mean?

A) the result is significant. The data suggest that parents who are more stressed have newborns that can sleep longer.
B) The result is significant. The data suggests that parents who are more stressed have newborns that sleep for shorter time.
C) The result is not significant. There is no evidence from this sample that suggests that parental stress and newborn sleep duration are related.
D) The result is significant. Parental stress causes newborns to sleep less.

Question 28

Which of the following is beyond the scope of a Pearson correlation?

A) calculating a linear relationship between two variables.
B) You can use it to determine how much one variable explains another variable.
C) Determine quadratic relationships between two variables
D) All of the above are correct.

Question 29

Truncated range is best described as when…

A) the population's variability is smaller than that of the sample's.
B) the sample's variability is smaller than that of the population's.
C) the sample's range is limited because the population is skeweed
D) the correlation coefficient overestimates the true population value.

Question 30

When a truncated range occurs, which of the following is a likely result?

A) The coefficient of determination is larger than the true value.
B) The correlation coefficient tends to over-estimate the true population value.
C) The correlation coefficient tends to under-estimated the true population value.
D) The correlation is more likely to be statistically significant.

Question 31

Which of the following is false of possible hidden variables in correlations?

A) Hidden variables can be the true cause of an outcome
B) The effects of hidden variables can always be controlled and fully accounted for after the data have been collected.
C) When hidden variables exist, they can either be directly or indirectly associated with the target variable.
D) All of the above are true.

Question 32

Which of the following best describes the relationship between statistics and causality?

A) The method and statistic play equal roles in determining whether causal relationships are established.
B) The statistic is the most important piece of information that determines whether causality can be inferred.
C) The method is more important than the statistic for determining a causal relationship
D) Causality has to do with sample size, but not the method or statistic. If you have the entire population's data, the results are always causal.

Question 33

Which of the following statements best describes the coefficient of determination?

A) It is a number constrained between -1.00 and 1.00
B) It is a number that varies from 0 to positive infinity
C) It is a number between 0 and 1.00
D) It is a percentage that represents the proportion of error variance explained by your independent variable.

Question 34

If you have a correlation coefficient of r = -.60, then your coefficient of determination should be…

A) 1.20
B) -.36
C) .36
D) -.30

Question 35

How many possible correlation coefficients could result in a coefficient of determination of exactly .49?

A) 1
B) 2
C) 3
D) 4

Question 36

If two variables are perfectly negatively correlated, their coefficient of determination would be…

A) -1.00
B) 1.00
C) 0.00
D) .50

Question 37

In a correlational dataset, if the coefficient of determination is positive, then…

A) The correlation from which the coefficient of determination was calculated is always positive
B) The correlation from which the coefficient of determination was calculated is always negative
C) There is insufficient information to know whether the correlation is positive or negative
D) The correlation from which the coefficient of determination was calculated can be positive or negative.

Question 38

In a regression analysis, b refers to

A) the y-axis intercept
B) the slope of the regression line
C) the predicted y value
D) the residuals

Question 39

A quadratic relationship differs from a linear relationship in that…

A) the line for a linear relationship changes direction once, but for quadratic relationships the line changes direction twice.
B) The two types of relationships have lines that change directions different amounts of times. But both should be analyzed by a Pearson correlation.
C) The line for a linear relationship never changes direction, but for quadratic relationships the line changes direction once.
D) None of the above.

Question 40

Which of the following demonstrates truncated range?

A) A researcher who is interested in IQ but only being able to recruit college students from elite private schools.
B) A researcher interested in personality, and collects data from the full range of socioeconomic status.
C) A researcher who is interested in the development of working memory across childhood during elementary, and recruits a large sample of 2^nd graders.
D) a and c only.

Question 41

When two variables have a coefficient of determination of .78, it suggests that…

A) One variable has a 78% chance of causing the other variable.
B) The two variables share 78% variance
C) The two variables have a correlation of .78
D) The two variables have a regression line that has a slope of .78

Question 42

Which one of the following variables cannot be dichotomized?

A) age of participant
B) reaction time in milliseconds
C) favorite type of soda
D) Number of coffees consumed per day

Question 43

The theoretical correlation of the population between two variables is denoted by…

A) r²
B) r
C) ρ
D) b

Question 44

Based on Galton's work, which of the following types of graphs would you expect to be most useful to plot a correlation between two variables?

A) scatterplot
B) bar chart
C) pie chart
D) histogram

Question 45

Even though researchers use regressions to predict outcomes, this does not mean that there is a(n) _____________.

A) correlation
B) causal effect
C) association
D) relationship

Question 46

Does money buy happiness? A research collected data on how much individuals spent for the holidays, and how satisfied they were with their holiday season, overall. Below are the data:
calculate the correlation

Question 47

Does money buy happiness? A research collected data on how much individuals spent for the holidays, and how satisfied they were with their holiday season, overall. Below are the data:
how many degrees of freedom are there for the test?

Question 48

Does money buy happiness? A research collected data on how much individuals spent for the holidays, and how satisfied they were with their holiday season, overall. Below are the data:
is the result significant at alpha = .05?

Question 49

Do high school grades predict college academic success? Below are data collected from some recent college graduates.
What is the correlation?

Question 50

Do high school grades predict college academic success? Below are data collected from some recent college graduates.
How many degrees of freedom are there?

Question 51

Do high school grades predict college academic success? Below are data collected from some recent college graduates.
is the test significant at the alpha = .05 leve? What about the .01 level?

Question 52

Do high school grades predict college academic success? Below are data collected from some recent college graduates.
What is the regression coefficient for the prediction?

Question 53

Do high school grades predict college academic success? Below are data collected from some recent college graduates.
What is the coefficient of determination?

Question 54

Is a person's degree of extraversion a predictor of romantic partners? Below are normed data on individuals' degree of extraversion and the number of partners they've had.
What is the correlation?

Question 55

Is a person's degree of extraversion a predictor of romantic partners? Below are normed data on individuals' degree of extraversion and the number of partners they've had.
How many degrees of freedom are there in this test?

Question 56

Is a person's degree of extraversion a predictor of romantic partners? Below are normed data on individuals' degree of extraversion and the number of partners they've had.
Is the correlation statistically significant at alpha = .05?

Question 57

Is a person's degree of extraversion a predictor of romantic partners? Below are normed data on individuals' degree of extraversion and the number of partners they've had.
identify a potential hidden variable that may explain the variable that the researcher failed to take into account

Question 58

Why is truncated range a problem when conducting correlational research?

Question 59

What is the phenomenon: Regression to the mean?