Module 2 Practice Questions
To estimate the sample sizes needed to detect a mean difference that is significantly different from zero you need which of the following: a. Effect size b. Standard deviation c. Critical value d. All of the above
Effect size
By what other name is the chi square goodness of fit? a. Two-sample chi square b. One-sample chi square c. Wilcoxon rank sum d. Chi square ANOVA e. None of the above
One-sample chi square
Which of the following indicates the purpose of the Kruskal-Wallis rank test? a. Test whether more than two groups are normally distributed. b. Test whether more than two groups have equal means. c. Test whether more than two groups are independent. d. Test whether more than two groups have equal medians.
Test whether more than two groups have equal medians.
Which type of statistical test assumes homoscedasticity? a. One-sample t-test b. Two-sample t-test c. Paired t-test d. One-sample z-test
Two-sample t-test
The more comparisons you test in an analysis, the higher the: a. Type I error. b. Type II error. c. Effect size. d. Posthoc error.
Type I error.
A more robust parametric alternative to the independent samples t test is the: a. matched pairs t test. b. one-way ANOVA. c. Welch's t test. d. Wilcoxon rank-sum test.
Welch's t test.
A researcher measured the same group of people's physiological reactions while watching horror films and compared them to when watching erotic films. The resulting data were skewed. What test should be used to analyse the data? a. Independent t-test b. Wilcoxon signed-rank test c. Dependent (related) t-test d. Mann-Whitney test
Wilcoxon signed-rank test
You're performing an experiment in which the variables can be sorted into 4 rows and 2 columns. Which post-hoc correction would be inappropriate for this type of variable? a. Pairwise Bonferroni correction b. Williams correction c. Yates correction d. All of the above would be appropriate
Yates correction
What is the alternative name for a repeated-measures t-test? a. A paired-samples t-test b. A Wilcoxon test c. Pearson's Product-Moment d. Chi-squared test of difference e. Spearman's Rho
A paired-samples t-test
Which of the following properties describe the chi-square distribution? a. Its shape depends on the degrees of freedom. b. As the degrees of freedom increase, the distribution becomes more symmetric. c. It is skewed. d. All of the above.
All of the above.
Assuming the assumptions of parametric tests are met, non-parametric tests, compared to their parametric counterparts: a. Are all of these. b. Are more conservative. c. Are less likely to accept the alternative hypothesis. d. Have less statistical power.
Are all of these
Which of the following tests is used to determine whether a set of observed frequencies differs from their expected frequencies? a. T b. F c. Chi-square d. Z
Chi-square
You want to determine whether there is a higher likelihood of cancer in women whose mothers smoked when they were pregnant. Your study has 200 women total. 80 have cancer and their mothers smoked. 20 have cancer and their mothers did not smoke. 60 do not have cancer and their mothers did not smoke. 40 do not have cancer and their mothers smoked. How would you calculate the odds that a woman will have cancer if her mother smoked when she was pregnant? a. (40*20)/( *60) b. (60*40)/(20*80) c. (80*60)/(40*20) d. (80/20)/(40/60)
(80*60)/(40*20)
Which of these statements is not true? a. A chi-square test of independence that is statistically significant shows a cause-and-effect relationship. b. A segmented bar chart is useful in observing when two variables might be associated. c. As the number of categories increases, the2χdistribution approaches the normal distribution. d. The chi-square tests involve categorical variables. e. The chi-square goodness-of-fit test is an extension of the z-test to more than two categories
A chi-square test of independence that is statistically significant shows a cause-and-effect relationship.
The Cochran-Mantel-Haenszel test is analogous to: a. A paired t-test for a measurement variable b. A Wilcoxon signed rank test for rank data c. A two-way ANOVA d. All of the above
All of the above
The t test for the difference between the means of two samples makes what assumption? a. Samples are randomly and independently drawn. b. Sample variances are equal. c. Populations are approximately normally distributed. d. All of the above.
All of the above
The central limit theorem suggests: a. The average of the sample means will itself be the population mean. b. The standard deviation of the sample means equals the standard error of the population mean. c. The sampling distribution of the means is also normally distributed even if the population is not. d. All of the above are true.
All of the above are true.
Which of the following is/are true? a. Parametric statistical tests involve data that are ratio or interval. b. Parametric statistical tests contain more assumptions that non-parametric tests. c. Non-parametric statistical tests are more suited to deal with data that are not normally distributed than parametric statistical tests. d. All of the above.
All of the above.
In testing a hypothesis about two population means, it the t distribution is used, which of the following assumptions is required? a. The standard deviations are not the same. b. Both populations are normally distributed. c. Both population means are the same. d. The sample sizes are equal.
Both populations are normally distributed.
If you are testing a hypothesis that two population proportions are the same, you should do which of the following? a. Calculate a pooled value for the sample proportion. b. Use a 0.05 level of significance. c. Use a sample proportion equal to 0.5. d. Average the two sample proportions.
Calculate a pooled value for the sample proportion.
You're performing an experiment that has 2 nominal variables and you want to know whether the proportion of one variable are the same for different values of the second variable. Normally, you would analyze the data using a ____________, but you also have an additional variable that reflects the different locations where your experiment took place. Therefore, you should use ______________ test instead. (n=1000) a. Fisher's exact test for independence; Cochran-Mantel-Haenszel b. Chi-Square test for independence; Repeated G-test c. Exact test; Repeated G-test d. Chi-Square test for independence; Cochran-Mantel-Haenszel
Chi-Square test for independence; Cochran-Mantel-Haenszel
A researcher asked 933 people what their favorite type of TV program was: news, documentary, soap or sports. They could only choose one answer. As such, the researcher had the number of people who chose each category of program. How should she analyse these data? a. t-test b. One-way analysis of variance c. Chi-square test d. Regression
Chi-square test
In which of the following cases would an independent-samples t-test not be appropriate? a. Comparing the performance of individuals who either consumed caffeine or did not b. Comparing younger and older adults' level of teamwork c. Comparing students attitude change between the start and end of their degree d. Comparing levels of aggression between 1 group who experienced high levels of excitation and 1 group who remained calm e. Comparing the attitudes of Eastern and Western individuals f. Comparing men and women's IQ on one occasion
Comparing students attitude change between the start and end of their degree
What is the definition of homoscedasticity? a. Equal standard error between samples b. Equal variance between samples c. Equal means between samples d. Equal medians between samples
Equal variance between samples
Which tests could be used if your expected cases were fewer than 5? a. Anova b. Paired or independent t-tests c. Pearson or Spearman's correlations d. Fisher's Exact Probability Test or Yates' Continuity Correction e. All of these
Fisher's Exact Probability Test or Yates' Continuity Correction
Chi-square is used to analyze: a. Scores b. Ranks c. Frequencies d. Any of these
Frequencies
What would a chi-square significance value of P > 0.05 suggest? a. That there is a significant difference between the sample and the population b. That there is a significant relationship between categorical variables c. That there is no significant difference between the sample and the population d. That there is a significant relationship between the sample and the population e. That there is no significant difference between categories
That there is no significant difference between the sample and the population
What property explains why there is a repeated G-test for goodness-of-fit, but not a repeated chi-square test for goodness-of-fit? a. The multiplicative nature of the G-statistic b. The exponential nature of the G-statistic c. The additive nature of the G-statistic d. The additive nature of the chi-square statistic
The additive nature of the G-statistic
When testing for differences between the means of two related populations, what is the null hypothesis? a. The difference between the two population means is equal to 0. b. The difference between the two population means is equal to 1. c. The difference between the two population means is greater than 0. d. The difference between the two population means is greater than 1.
The difference between the two population means is equal to 0.
After conducting a matched pairs t test, a researcher finds that the 95% confidence interval estimating the difference between the treatment group and the control group is 1.20 ± 1.5. Which of the following is definitely true? a. The difference is not significant. b. The difference is significant. c. The effect size is 3. d. The mean of the treatment group is 1.20.
The difference is not significant.
What is the definition of familywise error? a. The probability of a Type 1 error b. The probability of a Type 2 error c. The probability of at least one Type 1 error in a set of comparisons d. The probability of at least one Type 2 error in a set of comparisons
The probability of at least one Type 1 error in a set of comparisons
What is the definition of false discovery rate? a. The rate of Type 1 errors b. The rate of Type 2 errors c. The rate of false negatives d. The rate of return on investment
The rate of Type 1 errors
What is the purpose of a goodness-of-fit test? a. To identify significant effects b. To assess whether several categorical variables are related c. To find relationships in the data d. To assess whether there is a significant difference between a collection of categorical data e. To assess whether the central tendency, variability and distribution of sample is different from that of the population
To assess whether the central tendency, variability and distribution of sample is different from that of the population
How can you deal with low expected values? a. You can exclude outliers b. You can increase your sample size or combine categories c. You can add more of the same number d. You can transform your data e. You have to redo your experiment
You can increase your sample size or combine categories
Simpson's paradox refers to when: a. a trend disappears or reverses when groups of data are combined b. the ratio of L-null/L-alt increases as effect size decreases c. the ratio of L-null/L-alt gets smaller as the test statistic gets larger d. a trend persists despite pooling data
a trend disappears or reverses when groups of data are combined
After taking the natural log of the likelihood ratio, the ratio: a. gets bigger as the observed data get further from the null expectation b. gets smaller as the observed data get further from the null expectation c. gets bigger as the observed data get closer to the null expectation d. gets smaller as the observed data get closer to the null expectation
gets bigger as the observed data get further from the null expectation
Which equation represents the calculation of the correct test-statistic if the population variance is not known and you have a sample mean (as opposed to a unique value)? a. z=(x-μ)/σ b. z=(x ̅-μ)/(σ/√n) c. t=(x ̅-μ)/(s/√n) d. z=(x-x ̅)/s
t=(x ̅-μ)/(s/√n)
A local realtor is interested in testing if the median number of days a house is on the market before being sold exceeds 32 days. The following table shows the number of days that 7 houses were on the market. There is no evidence that the population is normally distributed. The appropriate statistical test is a. the Kruskal-Wallis test. b. the Wilcoxon matched-pairs signed rank test. c. the U-test. d. the Wilcoxon signed rank test.
the Wilcoxon signed rank test.
A matched pairs t-test compares means of ___________________ participants on ________________. a. different; a single measure b. the same; two different measures c. different; two different measures d. the same; a single measure
the same; two different measures
When comparing two population proportions, what is the null hypothesis if the alternate hypothesis is p1 > p2? a. H0: p1 > p2 b. H0: p1 not equal to p2 c. H0: p1 = p2 d. H0: p1 < p2
H0: p1 = p2
When comparing the difference between two population proportions, a pooled estimate of the population proportion can be used for two-tail tests where the null hypothesis assumes that the population proportions are equal. What is the alternate hypothesis? a. H1: p1 > p2 b. H1: p1 < p2 c. H1: p1 = p2 d. H1: p1 not equal to p2
H1: p1 not equal to p2
Which of the following is an assumption of the chi-square test? a. Random sampling and independence of observations b. All expected frequencies should be at least 5 for 2x2 tables. c. For larger tables, 4 out of 5 of expected call frequencies should be at least 5, and none should be zero. d. Homogeneity of variance across rows and columns of the table.
Homogeneity of variance across rows and columns of the table.
Which of the following is an assumption of the independent samples t test? a. Interval, ratio or ordinal data b. Random sampling c. The two populations from which the samples were drawn have equal variances d. The values in the two populations from which the samples are drawn are normally distributed
Interval, ratio or ordinal data
What is one disadvantage of a Bonferroni correction? a. It increases type 2 errors b. It decreases familywise error c. It decreases the power d. It increases the effect size
It decreases the power
A non-parametric alternative to the independent samples t test is the... a. chi-square test. b. Mann-Whitney U test. c. Wilcoxon signed-rank test. d. Bayesian analysis test.
Mann-Whitney U test.
If you are running a one-sample t-test and your data is symmetrical, bimodal, but not normal, should you run a Mann-Whitney U-test instead? a. Yes, because one-sample t-tests require data that is normal b. Yes, because the data is paired c. No, because the one-sample t-test is not sensitive to normality in this instance d. No, because a Wilcoxon sign rank test would be better
No, because the one-sample t-test is not sensitive to normality in this instance
What type of data do you need for a chi square test? a. Interval b. Ratio c. Nominal (categorical) d. Parametric e. Ordinal
Nominal (categorical)
If you are only interested in whether there is a difference between the two means and not the size of the difference, the best test to use would be: a. T-test b. Goodness-of-fit Test c. Sign test d. Z-test
Sign test
If you're running a two-sample t-test and one of your samples has a standard deviation that is twice as large as your other sample, what should you do? a. Log transformation b. Give up c. Perform a Welch's test d. Perform a post-hoc correction
Perform a Welch's test
After completing the statistical test in Question 1, you are concerned that you there will be a high probability of getting a p-value < 0.05 simply because your sample size was large, and you did multiple tests. Which of the following would be an INCORRECT way to correct this problem? a. Divide your alpha-value by your sample size to determine a new alpha-value b. Rank your p-values in ascending order (the smallest p-value has the lowest rank), calculate critical values for each p-value, find the smallest p-value that is bigger than its critical value, and determine that any p-values less than that one is significant. c. Divide your chi-square statistic by q, where q = 1 + (a^2 - 1)/6nv. d. All of the above would be acceptable corrections.
Rank your p-values in ascending order (the smallest p-value has the lowest rank), calculate critical values for each p-value, find the smallest p-value that is bigger than its critical value, and determine that any p-values less than that one is significant.
What are matched cases? a. Scores are compared across two distinct groups of participants b. A way of transforming data to make it more normally distributed c. Scores are obtained from a second group of participants who are matched on vital characteristics with the first group of participants d. Scores obtained from the same participants at time 1 and time 2 are standardized and compared e. Scores obtained from the same participants at time 1 and time 2 are not standardized but compared
Scores are obtained from a second group of participants who are matched on vital characteristics with the first group of participants
In which of the following cases could you use a paired-samples t-test? a. When comparing the same participants performance before and after training b. When assessing relationships between two groups c. When comparing men and women's scores d. When assessing goodness of fit e. When comparing two separate groups f. When assessing three groups or more
When comparing the same participants performance before and after training
Which of the following tests is used to determine whether a difference exists between two medians? a. Wilcoxon b. T c. F d. Kruskal-Wallis
Wilcoxon
Which statistical test would have this null hypothesis: the median difference between pairs of observations is zero? a. Two-sample t-test b. Paired t-test c. Wilcoxon Rank Sum d. Wilcoxon Sign Rank
Wilcoxon Sign Rank
A researcher was interested in stress levels of lecturers during lectures. She took the same group of 8 lecturers and measured their anxiety (out of 15) during a normal lecture and again in a lecture in which she had paid students to be disruptive and misbehave. The data were not normally distributed. Which test should she use to compare her experimental conditions? a. Paired samples t-test b. Mann-Whitney test c. Wilcoxon rank-sum test d. Wilcoxon signed-rank test
Wilcoxon signed-rank test
It has been claimed that drinking cranberry juice reduces blood pressure. Scientists investigate this question using 3 different experimental designs. i. A group of randomly sampled individuals are randomly assigned to two groups. One group is given cranberry juice everyday for 4 weeks, the other group is given a red drink everyday for 4 weeks. After 4 weeks the blood pressure for both groups is measured. ii. The blood pressure of a group of randomly sampled individuals was measured before the study. Then every day for 4 weeks they were given cranberry juice and their blood pressure was measured again after the study, the difference in blood pressure was analysed. iii. A group of randomly sampled individuals with known blood pressure was given cranberry juice everyday for 4 weeks. After 4 weeks their blood pressure was measured. The average blood pressure at the end of the study is compared with 140/90mmHg (which is the level where high blood pressure is diagnosed). Which test goes with each of the situations above? a. i = one sample proportion, ii = one-sample t-test, iii = two-sample t-test b. i = one sample t-test, ii =ANOVA, iii = one-sample t-test c. i =matched-pairs t-test, ii = independent two-sample t-test, iii = one-sample t-test d. i =independent two-samplFe t-test, ii = matched pairs t-test, iii = one-sample t-test e. i = independent two-sample t-test, ii = matched pairs t-test, iii = two-sample t-test
i =independent two-samplFe t-test, ii = matched pairs t-test, iii = one-sample t-test
The Wilcoxon matched-pairs signed rank test assumes that the 2 samples are: a. come from normally distributed populations. b. matched or paired. c. equal in size. d. All of the above.
matched or paired.
