Deck 14: Analyzing Your Data II: Specific Approaches

Question 1

What is not true in a parametric statistical analysis?

A) Rank order of data points are important
B) The data should be normally distributed
C) The variance is taken into account
D) Sample size does not influence probability of a significant result

Answer 1

D

Question 2

A researcher is interested in the effect of maternal warmth on academic achievement. He devised a scale that measures warmth and uses student GPA has a measure of academic achievement. He finds that the correlation between the two variables is .60 and has a regression coefficient of .15. Which of the following statements is most accurate?

A) The researcher can conclude that a one unit increase in maternal warmth will cause a .15 decrease in GPA.
B) Maternal warmth explains 36% of the sample's variation in GPA
C) Students' GPA are lower for mothers who exhibit more warmth
D) The researcher can conclude that a one unit increase in maternal warmth is associated with a .60 increase in GPA.

Answer 2

B

Question 3

A researcher analyzed his data using an independent samples t test and found that the 95% confidence interval of the difference between two means ranges from -0.54 to -0.20. Which of the following is true?

A) The means for both groups must be negative.
B) Both groups have a negative association.
C) The means for both groups are not equal to one another.
D) Both groups are different from each other 95% of the time.

Answer 3

C

Question 4

The non-parametric alternative of a matched-pairs t test is the:

A) Mann-Whitney U test.
B) Bayesian approach.
C) Wilcoxon signed-rank test.
D) point-biserial correlation.

Question 5

A researcher conducted a study looking at how participants' alertness influenced their ability to complete an academic task with, or without coffee. All participants were randomly assigned to either stay up all night or get rest, and subsequently randomly assigned to drink regular coffee or decaffeinated coffee. He finds that there was an interaction between these two factors on the outcome variable. Which of the following is definitely true?

A) With the information provided, there must either be a significant main effect of alertness OR caffeine
B) With the information provided, caffeine and alertness must both be significant main effects
C) With the information provided, neither caffeine nor alertness must necessarily be significant main effects.
D) With the information provided, caffeine and alertness are either both significant main effects, or both not significant main effects.

Question 6

The non-parametric alternative to the independent groups one-way ANOVA is the:

A) Wilcoxon signed-ranked test
B) Kruskal-Wallis test
C) Welch's t-test
D) Wilcoxon rank-sum test

Question 7

Which of the following is not a common feature shared by the chi-square test and the independent samples t-test?

A) They both have the assumption of randomly sampling and independence of observations.
B) They both give you a p-value.
C) They both involve taking the degrees of freedom into account.
D) They both can be applied to contingency tables.

Question 8

What is an additional concern in terms of assumptions that one must check when conducting repeated measures design that is not a concern in a between subjects design?

A) homogeneity of variance
B) normality
C) independence
D) sphercity

Question 9

Which of the following regarding ANOVA and regression are true?

A) Whereas ANOVA is utilized for categorical independent variables, regression is only used for continuous independent variables
B) Any data that is analyzed with ANOVA can be analyzed with regression yielding the same outcomes.
C) ANOVA and regression have little in common. ANOVA is only used to analyze up to two factors at once, whereas regression analyses can include a much larger number of variables.
D) ANOVA can be used to analyze experimental designs that allow for causal claims, whereas regression is only used for correlational analyses that do not afford causal inferences.

Question 10

Which of the following is true?

A) Independent samples t test is used when your conditions are run within subjects.
B) Effect sizes can be computed for both between subject designs and within subject designs.
C) If you have more than one comparison group and all your conditions are run between subjects, then the test to use is a mixed ANOVA.
D) In a between subjects design with two groups and one factor, you should only use the t test and not a one-way ANOVA.

Question 11

Spearman's rho is more suitable than Pearson's r as a statistic when:

A) the measurement scale is ratio.
B) the measurement scale is ordinal.
C) the measurement scale is categorical.
D) the measurement scale is interval.

Question 12

An assumption in linear regression is the independence of the:

A) differences between predicted and actual dependent variable values.
B) associations between the independent variables and the dependent variable.
C) differences between the elements in the regression equation.
D) associations between the y-intercept and the mean of the dependent variable values.

Question 13

Charlotte is investigating the influence of multiple factors such as age, gender and income on how much money an individual spends on grocery shopping every month. Which of the following statistical tests best suits her study design?

A) Linear regression
B) Chi-square test
C) Pearson correlation
D) Spearman rank order correlation

Question 14

Which of the following do outliers not affect?

A) Standard deviations
B) Probability of Type I error
C) Confidence intervals
D) None of the above

Question 15

In Bayesian data analysis, the objective prior is used when:

A) researchers have a clear idea of what the prior distribution is.
B) the odds of obtaining every possible outcome are objectively known.
C) researchers have little idea of what the prior distribution is.
D) researchers are concerned with reducing Type I error.

Question 16

Pauline is testing the claim that students from her university have better spatial memory than the average student in the country. She measures the average spatial memory of 50 students and compares it to the national average. What test is she using and what are the degrees of freedom of this test?

A) Independent samples t test; 49
B) One sample t test; 48
C) One sample t test; 49
D) Independent samples t test; 48

Question 17

According to your textbook, violation of sphericity when using a repeated measures design is likely to affect which of the following?

A) The probability of committing a Type I error.
B) The probability of committing a Type II error.
C) The probability of increasing findings that reflect the null hypothesis.
D) The probability of having skewed data.

Question 18

After conducting an independent samples t-test, a researcher finds that the 95% confidence interval estimating the difference between the treatment group and the control group is 2.37 ± 1.85. Which of the following is definitely true?

A) The difference between groups is not significant.
B) The difference between groups is significant.
C) The effect size of the comparison between groups is 3.7.
D) The mean of the treatment group is 2.37.

Question 19

Sofia is comparing the effectiveness of a new therapy on depression with a more traditional form of treatment. She compares the prevalence of depression symptoms in 15 patients who have undergone the new therapy with the symptoms of 15 other patients who have undergone the traditional treatment. Which of the following test is most suited to the design of her study?

A) Mixed ANOVA
B) Independent samples t-test
C) Related samples t-test
D) Linear regression

Question 20

Which of the following is not an assumption of the independent samples t-test?

A) The data is interval or ratio.
B) Observations are unrelated.
C) The two populations being compared have equal variances.
D) The effect size is not zero.

Question 21

Which of the following is true?

A) When the p-value is less than 0.05 for an independent samples t test, the 95% confidence interval of the difference between groups will cross zero.
B) When the p-value is less than 0.05 for a one-sample t test, it means that the assumptions of the test have been met.
C) When the p-value is larger than 0.05 for any test, it means that the effect size is zero.
D) When the p value is larger than 0.05 for an independent samples t test, it means that there are no significant differences between groups at the level of alpha = .05.

Question 22

When analyzing the data from a study with 100 participants using a one-way between subjects ANOVA, a researcher finds that the total sum of squares is 565 and the within-subjects sum of squares is 120. Which of the following is true?

A) The between-subjects sum of squares is 56.5.
B) The between-subjects sum of squares is 445.
C) The between subjects sum of squares is 685.
D) There is insufficient information to calculate the between-subjects sum of squares.

Question 23

The table below presents the results of a between subjects one-way ANOVA experiment. refer to this table.

-What is the value of F?

A) 8
B) 3
C) 30
D) 3.68

Question 24

The table below presents the results of a between subjects one-way ANOVA experiment. refer to this table.

-What is the value of the within subjects degrees of freedom?

A) 24
B) 1
C) 27
D) 29

Question 25

The table below presents the results of a between subjects one-way ANOVA experiment. refer to this table.

-What is the total sum of squares?

A) 1000
B) 660
C) 290
D) 80

Question 26

The number of degrees of freedom (df) is the number of experimental data that have contributed to the test statistic.

Question 27

The independent samples t test is used to compare the means of unrelated two samples coming from populations that have vastly different variances.

Question 28

The Mann-Whitney U test is a statistically more powerful alternative to the independent samples t-test.

Question 29

The sums of squares in an ANOVA test measure the amount of variability in the data.

Question 30

The omnibus null hypothesis tested by one-way ANOVA is that the population means for all the groups differ at least once.

Question 31

According to Cohen (1988), obtaining an η² of 0.5 indicates a large effect.

Question 32

The η² from a between-subjects one-way ANOVA tells you whether or not your results are due to chance.

Question 33

Spearman's rho is a more conservative test and requires more assumptions than Pearson's r.

Question 34

Linear regression can be used to analyze ANOVA designs with multiple factors (i.e., independent variables).

Question 35

Linear regression is suitable for examining all types of relationships between variables, as long as the data are on an interval or ratio scale.

Question 36

Random sampling and independence of observations are assumptions of many statistical tests that should be taken into account while designing the research.

Question 37

To check whether the assumption of equal variances across the different groups holds in order to conduct an ANOVA test, researchers refer to the Levene's test.

Question 38

The Welch's t test is a parametric test that has fewer assumptions than an independent groups t-test.

Question 39

The purpose of objective priors in Bayesian analysis is to provide researchers with a prior distribution that makes no assumptions whatsoever.

Question 40

Bayesian analysis requires the underlying population to be normally distributed.

Question 41

Jessica wants to analyze her data using an independent samples t test. However, she is unsure about whether her data violates the assumption of equal variances across groups. List two types of tests or techniques that Jessica might use in place of an independent samples t test. In your answer, specify whether each solution is a parametric or non-parametric method.

Question 42

List four assumptions of a repeated measures one-way ANOVA. Explain what is meant by sphericity.

Question 43

Describe the three main conceptual components to a Bayesian analysis.

Question 44

List two advantages and two disadvantages of using the Bayesian approach to data analysis.

Question 45

George plans to analyze his data with a two-way ANOVA but he is concerned about the presence of outliers in his data set. Is his concern warranted? Why?