Exam 2

Question 1

Which statement is NOT true regarding the slope of a regression equation?
A) The slope quantifies the steepness of the line
B) If the slope is negative, Y tends to decrease as X tends to decrease
C) If the slope is positive, Y tends to increase as X tends to increase
D) None of the above, all are true

Answer 1

B) If the slope is negative, Y tends to decrease as X tends to decrease

Question 2

The scale of the correlation coefficient is:
A) Nominal
B) Interval
C) Ordinal
D) Ratio

Answer 2

C) Ordinal

Question 3

A fellow researcher tells you that she got a p calculated of .04 from an inferential test she conducted. She asks you if her results are statistically significant. The most informed response would be:
A) Yes, because the probability of obtaining your sample statistic was less than .05
B) Yes, because your alpha level was greater than your obtained p calculated
C) No, because the probability of obtaining your sample statistic was greater than .05
D) No, because your alpha level was greater than your obtained p calculated
E) It depends, what did you set your probability of Type I error to be?

Answer 3

E) It depends, what did you set your probability of Type I error to be?

Question 4

Statistical significance testing (null hypothesis tests) depends on the assumption that:
A) You have a large sample size
B) The null hypothesis is true in the population
C) You have a low risk if Type I error
D) The alternative hypothesis cannot be proven true
E) None of the above

Answer 4

B) The null hypothesis is true in the population

Question 5

A statistically significant result means that, given the assumptions of the test, you have a(n):
A) Replicable result
B) Important result 
C) Unlikely result
D) Unimportant result

Answer 5

C) Unlikely result

Question 6

In the process of testing a null hypothesis, which of the following should precede the others.
A) Computing the test statistic
B) Deciding whether to reject Ho
C) Determining your conclusion regarding group differences
D) Establishing the criterion for rejecting

Answer 6

D) Establishing the criterion for rejecting

Question 7

Which of the following is not true about Pearson r and Spearman rho?
A) r ranges from -1 to +1
B) When Spearman rho = 1.00, then Pearson r will always be 1.00
C) When Pearson r = 1.00, then Spearman rho will always be 1.00
D) r and rho cannot be compared without more information

Answer 7

B) When Spearman rho = 1.00, then Pearson r will always be 1.00

Question 8

Which alpha level below corresponds to the most risk of Type I error?
A) alpha = .10 
B) alpha = .05 
C) alpha = .01 
D) alpha = .001

Answer 8

A) alpha = .10

Question 9

Which of the following is an example of an alternative hypothesis?
A) Exercising 30 minutes a day for 2 weeks does not decrease heart rate by 10%
B) Taking 500mg of ibuprofen within 20 minutes of onset will decrease headache pain
C) Studying more than 5 hours for a statistics final does not impact your grade
D) There is no difference between the mean self-esteem scores of mean and women athletes at UNT

Answer 9

B) Taking 500mg of ibuprofen within 20 minutes of onset will decrease headache pain

Question 10

The primary goal of inferential statistics is to:
A) Describe characteristics of samples based on population parameters
B) Suggest characteristics of populations based on sample statistics
C) Find exact values for population parameters with our sample statistics]
D) Find exact values for sample statistics given estimated population parameters
E) Drive graduate students “batty”

Answer 10

B) Suggest characteristics of populations based on sample statistics

Question 11

Standard error is the standard deviation of a sampling distribution. The larger the standard error, the greater risk of \_\_\_\_\_\_\_\_\_\_\_\_\_ error.
A) Sampling 
B) Type I 
C) Type II 
D) Type III 
E) None of the above

Answer 11

C) Type II

Question 12

Given r = .8 and a p calculated of .6, when we test the null hypothesis that r = 0, we can say that:
A) There is an 80% probability of obtaining at least r = .8 from any random sample, if the r should have been 0
B) There is a 60% probability of obtaining at least r = .8 from any random sample, if the r should have been 0
C) There is a 60% probability of not obtaining at least r = .8 from any random sample, if the r should have been 0
D) There is an 80% probability of not obtaining at least r = .8 from any random sample, if the r should have been 0

Answer 12

B) There is a 60% probability of obtaining at least r = .8 from any random sample, if the r should have been 0

Question 13

Which of the following is true regarding weak correlation coefficients?
A) They indicate a slight causation between variables
B) They are found only with Pearson r coefficients
C) They may be the result of restricted range for one of the variables
D) They are rarely found in social science research
E) They account for large amounts of dependent variable variability

Answer 13

C) They may be the result of restricted range for one of the variables

Question 14

A Pearson r of .70 is how much stronger than a Pearson r of .00 in terms of shared variance between the two variables?
A) 30%
B) 40%
C) 49%
D) 70%
E) None of the above

Answer 14

C) 49%

Question 15

In hypothesis testing, if we found a statistically significant difference between the means of two groups, then we:
Correct!
A) Rejected the null
B) Failed to reject the null
C) Committed Type I error
D) Committed Type II error
E) None of the above

Answer 15

A) Rejected the null

Question 16

When choosing a non-directional hypothesis, we:
A) Are choosing a one-tailed test
B) Decrease the likelihood of Type II error
C) Increase the likelihood of Type I error
D) Decrease the likelihood of statistical significance
E) None of the above

Answer 16

D) Decrease the likelihood of statistical significance

Question 17

As a special education researcher, you know that easy distractibility and diminished attention at school have been strongly correlated with low scores on certain tests of sensory integration. A child is found to have trouble paying attention in his classes. The school counselor asks for sensory integration therapy to remediate the attention problem. Should the therapist provide therapy right away?
A) Yes, because the strong correlation proves that therapy which improves sensory integration will definitely improve attention
B) No, because therapy caseloads in public schools are high and this child s problems are relatively minor
C) No, because a strong correlation doesn’t mean that one condition necessarily causes the other; more investigation is needed to determine the possible causes of the child s distractibility
D) Yes, because the school administrators are in favor of the therapy

Answer 17

C) No, because a strong correlation doesn’t mean that one condition necessarily causes the other; more investigation is needed to determine the possible causes of the child s distractibility

Question 18

Dr. Marcela conducted a study on SAT and GRE scores and calculated the Pearson r correlation between these scores for a sample of graduate students. Which correlation coefficient is least likely to reflect the relationship between these variables?
A) -1.0
B) -0.1
C) 1.4
D) 1.0
E) None of the above, all are equally plausible

Answer 18

C) 1.4

Question 19

The bivariate reference point that is analogous to the univariate mean is called the:
A) Covariance
B) Correlation
C) Coordinate
D) Centroid

Answer 19

D) Centroid

Question 20

In a regression linear equation for one predictor and a dependent variable, in which of the following scenarios will the a weight always be 0?
A) When the SDs of both variables are the same
B) When the means of both variables are the same
C) When the SDs of both variables are 0
D) When the variables are in Z score form
E) None of the above

Answer 20

D) When the variables are in Z score form