Deck 18: Inferential Statistics

Question 1

In a recent study that tested differences in metacognition between those with high ability and those with low ability, Teo concluded that there were not differences between these groups. In fact, however, the low ability students were actually significantly better at monitoring their learning. Teo's conclusion represents

A) a Type I error.
B) a Type II error.
C) both a Type I and a Type II error.
D) a correct decision.

Answer 1

B

Question 2

In her treatment group, Kenesha had students with developmental disabilities swim daily for 20 minutes. In her findings, she reported that at the conclusion of the study, those in the treatment condition had greater self-confidence than those in the no-swimming control group. There were actual differences between groups. Kenesha's conclusion represents

A) a Type I error.
B) a Type II error.
C) both a Type I and Type II error.
D) a correct decision.

Answer 2

D

Question 3

As the population standard deviation increases

A) sampling error decreases.
B) sample mean decreases.
C) standard error increases.
D) population mean increases.

Answer 3

C

Question 4

Mike concluded that there were no differences in self-confidence between those who had received his intervention and those in the control group. However, there were true differences, and Mike's intervention did actually work. Mike's conclusion represents

A) a Type I error.
B) a Type II error
C) both a Type I and Type II error.
D) a Type III error.

Question 5

Which of the following represents a null hypothesis regarding the differences between two groups learning biology content?

A) There are differences in a practicum exam between those who learn dissection in the actual lab and those who learn dissection by computer simulation.
B) There are no differences in a practicum exam between those who learn dissection in the actual lab and those who learn dissection by computer simulation.
C) Those students who learn dissection in the lab will perform better on the practicum exam than those who learn dissection via the computer simulation.
D) Those students who learn dissection via the computer simulation will perform better on the practicum exam than those who learn dissection in the lab.

Question 6

If a researcher rejects the null hypothesis she is concluding that

A) there are real differences between groups.
B) there are no real differences between groups.
C) differences between groups are due to chance.
D) differences between groups are due to measurement error.

Question 7

When a researcher concludes that there are no differences between the treatment and control groups in a study, and in fact, there were no differences between the conditions. The researcher is illustrating

A) Type I error.
B) Type II error.
C) both Type I and Type II error.
D) a correct decision.

Question 8

The smaller the standard error the

A) more accurate the sample means are as an estimate of the population mean.
B) more likely differences between groups are due to chance.
C) less likely that the statistic represents the parameter.
D) less accurate the subject's response is on a dependent measure.

Question 9

Darlene concludes in her thesis that there are significant differences in reading motivation between those children whose parents read to them daily and those who did not. There were, however, no actual differences between these groups. Darlene's conclusion represents

A) a Type I error.
B) a Type II error
C) both a Type I and a Type II
D) a correct decision.

Question 10

Anan conducted an experimental study in which students were randomly assigned to either 10 minutes of read aloud, 10 minutes of silent reading, 10 minutes of writing, or 10 minutes of free time. She concluded that there were significant differences among groups with the read aloud group outperforming others on a standardized reading assessment. However, there were not actual differences among groups. Anan conclusion represents

A) a Type I error.
B) a Type II error.
C) both a Type I and a Type II error.
D) a correct decision.

Question 11

Researchers determine the probability level

A) prior to the execution of the study and any data collection.
B) after the data are collected by prior to analysis
C) after analysis but prior to interpretation
D) after the data are analyzed and interpreted but before writing it up.

Question 12

As standard error of the mean increases

A) population mean decreases.
B) sample mean increases.
C) sampling error increases.
D) sample mean decreases.

Question 13

Inferential statistics are generally concerned with

A) how well the sample represents the population.
B) if the correct statistical test is used for analysis.
C) the internal validity of the research design.
D) the consequential validity of the dependent measure.

Question 14

When a researcher is concerned about committing a Type I error and wants a stringent level of significance, which of the following significance levels is best to use to reject or retain the null hypothesis?

A) .0001
B) .01
C) .05
D) .10

Question 15

Given a set of scores with a mean of 25, standard deviation of 4 and a sample size of 50, calculate the standard error of the mean.

A) .36
B) .57
C) 1.75
D) 2.42

Question 16

As the probability of committing a Type I error increases

A) the probability of committing a Type II error decreases.
B) the probability of committing a Type II error generally increases.
C) the probability of committing a Type II error may increase or decrease.
D) the probability of committing a Type II error will remain constant.

Question 17

Given a set of scores with a mean of 80, a standard deviation of 3, and a sample size of 82, calculate the standard error of the mean.

A) .27
B) .33
C) .68
D) .80

Question 18

Which of the following does one need to know to calculate the standard error of the mean?

A) Mean and sample size.
B) Mean and standard deviation.
C) Standard deviation and sample size.
D) Mean, standard deviation, and sample size.

Question 19

If a null hypothesis is rejected it means

A) the treatment causes significant differences.
B) there are no differences between treatments.
C) any differences between groups are due to sampling error.
D) the differences found between groups are not due to chance.

Question 20

Which of the following represents a null hypothesis regarding differences in SAT score by demographic characteristics?

A) There are no differences in SAT scores among students of different race and ethnicities.
B) Hispanic students score lower than any other students on the SAT.
C) Hispanic students score higher than any other students on the SAT.
D) There are significant differences in SAT among students of different race and ethnicities.

Question 21

A nonparametric test should be used if

A) the sample size is small.
B) the level of measurement is nominal.
C) the scores are normally distributed.
D) the concern for a Type II error is great.

Question 22

The t test for nonindependent samples is when

A) the two groups tested are randomly formed samples.
B) the two groups tested are initially the same on the dependent variable.
C) the two groups are formed by preexisting samples.
D) matching has been done or in a pre-post design.

Question 23

Independent samples refer to samples

A) selected from different populations.
B) selected from the same population.
C) known to be different on extraneous variables.
D) selected based upon a stratified sample.

Question 24

The statistical test that allows us to determine the degree to which variables are related is

A) ANCOVA.
B) MANOVA.
C) Multiple regression.
D) Chi Square.

Question 25

When one does follow-up tests to determine significant differences between groups in a comparison between means test, one test to use is the

A) t test for nonindependent samples.
B) Scheffe test.
C) Chi Square.
D) Pearson r.

Question 26

In determining the test of significance to select, one must consider all of the following EXCEPT

A) number of groups being compared.
B) number of independent variables.
C) the effect sizes needed.
D) scale of measurement represented by data.

Question 27

Shani is testing whether her intervention designed to increase self-reported tolerance for others in aggressive youth is effective. She administers her scale before and after the intervention and concludes that her intervention is effective. Which analysis did Shani most likely use?

A) The t test for independent samples
B) The t test for nonindependent samples
C) ANCOVA
D) ANOVA

Question 28

One benefit of conducting a two-tailed test is that

A) it allows for a greater region of rejection than a one tailed test.
B) it allows for differences in groups be in either direction.
C) it is easier to obtain significant differences between groups.
D) it is easier to reject the null hypothesis than for a one tailed test.

Question 29

Misha is comparing different types of typing instruction. In one group, he used a classroom-based instructional method, and for the other group, he used a computer-based training program. After a six-week treatment, he assessed differences in their typing ability. Of the following, which is the appropriate test of significance for Misha to use assuming random assignment and no pretreatment differences?

A) The t test for independent samples
B) The t test for nonindependent samples
C) Analysis of difference scores
D) ANOVA

Question 30

The first decision in selection of an appropriate test of significance is

A) to examine sample size.
B) to determine if a parametric or nonparametric test must be selected.
C) to consider the effect size you need.
D) to decide if your level of significance is appropriate.

Question 31

The level of probability selected determines the

A) probability of committing a Type I error.
B) probability of committing a Type II error.
C) probability of committing both a Type I and Type II error.
D) neither Type I or Type II error is related to selected probability.

Question 32

Rose is sure that any significant differences in her reading intervention study will favor those children who were in her treatment condition. She therefore will not consider rejecting the null hypothesis that the groups are different in favor of her control condition. Rose will then

A) use a non-parametric test of significance.
B) apply a two-tailed test of significance.
C) apply a one-tailed test of significance.
D) use descriptive findings only to examine her data.

Question 33

Parametric tests require all of the following assumptions EXCEPT

A) that there are an equal number of participants per group.
B) that data are interval or ratio scale of measurement.
C) that the selection of participants is independent.
D) that the variable measured is normally distributed.

Question 34

The test of significance that determines if two group means differ more than would be expected by chance is the

A) t test.
B) chi square.
C) multiple regression.
D) analysis of gain scores.

Question 35

Vincente intends to examine differences between three groups of district curriculum coordinators that received either no formal training, a web-based training, or a cooperative group experience as instruction for the core standards. He tested participants' knowledge of the standards both before treatment and then after treatment. Given his design, which of the following analysis should Vincente conduct?

A) The t test for independent samples
B) The t test for nonindependent samples
C) ANCOVA
D) ANOVA

Question 36

Given a set of scores with a mean of 80, standard deviation of 8 and a sample size of 200, calculate the standard error of the mean.

A) .57
B) .85
C) 1.18
D) 1.62

Question 37

If an observed F statistic (4.80) with 1 degree of freedom within and 12 degrees of freedom between and a significance level of p =.04 and the table value is 4.75 what can the researcher conclude?

A) He has proven there are differences between groups.
B) There are no differences between groups.
C) Differences between groups are due to chance.
D) Differences between groups are greater than those expected by chance.

Question 38

Nelson's recent study tested differences in problem-solving ability based upon science ability, treatment (instruction, no instruction), and gender. Of the following, which test should Nelson employ?

A) Chi Square
B) t test
C) ANOVA
D) Multiple regression

Question 39

Hal administers a self-esteem measure to his sample before the sessions begin and after a semester of the Saturday Esteem Program, he measures their esteem again. Which type of analysis is Hal likely to use to test for significant differences in his group?

A) The t test for independent samples
B) The t test for nonindependent samples
C) Analysis of difference scores
D) Analysis of covariance

Question 40

Cristos has collected data to explore differences in opinions about diversity training in the workplace based upon geographic location in the county. The first step he should take in selecting the appropriate test of significance is to

A) conduct a power analysis.
B) consider the effect size he needs.
C) decide his level of significance.
D) decide if a parametric or nonparametric test will be used.

Question 41

Katie questions whether there are significant differences in her study but her significance level is unable to detect these differences. What Type of error is Katie most concerned that she is committing?

A) Type I
B) Type II
C) Type III
D) Type IV

Question 42

A researcher is concerned about not being able to detect true differences. From the following options, how would you advise the researcher regarding her analyses?

A) Change the probability level to p= .001
B) Evaluate and conduct the study with a new sample
C) Reanalyze with a two-tailed analyses
D) Use a non-parametric significance test

Question 43

Garrett wants to see if there are gender differences in the restaurant preferences of high school students in his town. His data is nominal in nature. Which of the following tests is he likely to employ?

A) t test for nonindependent samples
B) t test for independent samples
C) Chi-square
D) ANOVA

Question 44

Given a set of scores with a mean of 80 and a standard deviation of 10 and a sample size of 100, calculate the standard error of the mean.

A) .48
B) .76
C) 1.00
D) 2.24

Question 45

The results of an ANOVA to test the difference in means between three treatment conditions indicate significant differences between conditions. Which of the following tests might a researcher use to determine where these differences are found?

A) Spearman rho
B) t test of nonindependent samples
C) ANCOVA
D) Scheffe test

Question 46

Molly has conducted a study that examines whether exposure to computers increases eyestrain in children. She randomly assigns children to one of four treatment conditions dependent upon computer exposure levels. The children's eyestrain is then tested. Which of the following tests of significance is Molly likely to use?

A) t test
B) Chi square
C) ANOVA
D) Analysis of gain scores

Question 47

In his study of effectiveness of driver's education programs, Cameron administers a practice items from a drivers' test. Participants are randomly assigned to one of two treatments, either a tutorial that narrates content, or an equivalent amount of time to practice items with feedback. After the treatment, Cameron again assesses students' knowledge with a drivers' test. He set his alpha at p = .05, and reported that significant differences indicated that those in the practice with feedback condition outperformed those in the tutorial condition. Which of the following were appropriate for Cameron to use to analyze his data?

A) t test
B) ANCOVA
C) Chi Square
D) Pearson r

Question 48

Given a set of scores in a normal distribution with a mean of 80, standard deviation of 4, and sample size of 100. If the sample were increased to 200 participants would we expect to happen to the standard error?

A) The standard error would be unaffected
B) The standard error would decrease
C) The standard error would increase
D) The standard error would double

Question 49

Which of the following represents a null hypothesis regarding the differences between two treatment groups of students exposed to behavior management strategies for social behavior?

A) There are differences in observed prosocial behavior between those exposed to a graphing strategy and those exposed to a cognitive self-reflection strategy.
B) Those students who are given a graphing strategy will display more prosocial behavior than those given a cognitive self-reflection strategy.
C) There are no differences in prosocial behavior between those exposed to a graphing strategy and those exposed to a cognitive self-reflection strategy.
D) Those students who are given a cognitive self-reflection strategy will display more prosocial behavior than those given a graphing strategy.

Question 50

In her study of student athlete's achievement motivation, Katie administers a self-report achievement motivation scale before she starts a new motivation enhancement intervention. After the intervention, Katie again assesses students' achievement motivation. She set her alpha at p = .05, and reported that there were no significant findings in her study. Which statistical analysis was appropriate for Katie to use to determine if there were treatment differences in her study?

A) Pearson r
B) ANOVA
C) Spearman rho
D) t test

Question 51

Saud examined differences in reading ability between those who were exposed to parent read aloud and those who were exposed to sibling read aloud. To calculate his t test to compare groups, Saud needed to determine his degrees of freedom. To do so he will need to know not only that he has two groups but also he will need to know which of the following pieces of information?

A) The number of children in the study.
B) The number of parents in the study.
C) The mean group performance.
D) The mean number of treatment sessions.

Question 52

In his study of effectiveness of driver's education programs, Cameron administers a practice items from a drivers' test. Participants are randomly assigned to one of two treatments, either a tutorial that narrates content, or an equivalent amount of time to practice items with feedback. Which of the following best represents the type of research Cameron conducted?

A) Correlational
B) Experimental
C) Descriptive
D) Narrative