Deck 17: Overview and Extension: Statistical Tests for More Complex Designs

Question 1

Given the information above,assume that the sample population was not normally distributed what statistical test would be appropriate?

A)Wilcoxon Signed-Rank test
B)Mann-Whitney U test
C)Spearman Rank order
D)Kruskal-Wallis Test
E)none of these

Answer 1

Wilcoxon Signed-Rank test

Question 2

The nature of the relationship in the independent groups t-test is determined by examining:

A)eta-squared
B)omega squared
C)sample means
D)population means
E)none of these

Answer 2

sample means

Question 3

Dr.Johnson sets his alpha level at .05 and uses a directional test,his degrees of freedom is 60 what is his critical t?

A)2.00
B)2.021
C)1.671
D)1.684
E)none of these

Answer 3

1.671

Question 4

The Independent groups t-test assumes _____.

A)normal distribution
B)Dependent V.ratio or interval
C)between subjects design
D)All of these
E)None of these

Question 5

Dr.Smith is studying the relationship between the use of a Ipod (portable headphones)and the ability to perceive sounds that have high pitches.She has 2 groups of subjects: one group that does not listen to earphones (N=20)and a group that does listen to earphones (frequently)(N=20).After collecting her data,she has asked you to analyze her results.What statistical test will you use?

A)Correlate t-test
B)Independent t-test
C)One-way ANOVA (between subjects)
D)ANOVA (repeated measures)
E)Two-way ANOVA (between subjects)

Question 6

Dr.Nan is studying the effects of Ginseng on recall.He has 2 groups of subjects, (20 subjects in each group).Subjects are given Ginseng or Placebo and then asked to recall a list of words they previously studied.However,to avoid confounds what should Dr.Nan do?

A)Counterbalance placebo & Ginseng
B)Match groups on an IQ test and assign to groups
C)Randomly assign subjects to groups
D)All of these
E)None of these

Question 7

In the study above what type of statistical test,should Dr.Nan use (if subjects have been matched on age or IQ)?

A)Independent t-test
B)Correlated t-test
C)One-way ANOVA (between subjects design)
D)One-way ANOVA (repeated measures)
E)none of these

Question 8

The calculation of eta²_____when the statistical test is not significant.

A)is not necessary
B)is misleading
C)can be useful
D)redundant
E)None of these

Question 9

If you have a small sample size what type of statistics should you use?

A)parametric statistics
B)nonparametric statistics
C)descriptive statistics
D)convert to z-scores
E)None of these

Question 10

The power of a statistical test is influenced by_____.

A)the strength of the relationship between 2 variables
B)sample size
C)alpha level
D)both a & c
E)All of these are correct

Question 11

When sample sizes are small,power will tend to be_____,everything else being equal.

A)low
B)high
C)moderate
D)there is no affect
E)none of these

Question 12

A statistical test led Dr.Jones to incorrectly decide that his independent variable had no affect on the dependent variable,when in fact it did.Dr.Jones committed a(n)

A)Type I error
B)Type II error
C)no error
D)1-a
E)none of these

Question 13

In the study above the independent variable is_____.

A)the perceived sounds
B)the use of a ipod
C)recall for perceived sounds
D)all of these
E)none of these

Question 14

Dr.Belair is studying the effects of noise exposure on the occurrence of headaches.He exposes 20 subjects to a constant noise in their work environment and asks them to record their headaches.The next week he uses the same subjects without exposing them to noise and measures the number of headaches the subject's record.What statistical test should Dr.Belair use to analyze his data?

A)Independent t-test
B)Correlated t-test
C)One-way ANOVA (between subjects design)
D)One-way ANOVA (repeated measures)
E)None of these

Question 15

Statistics that allow you to evaluate your data and generalize to the population are called_____.

A)inferential statistics
B)descriptive statistics
C)reliable
D)statistical power
E)None of these

Question 16

Before a researcher decides what statistical test,he or she should use to analyze their data they should consider the factors below _______.

A)The number of independent variables manipulated
B)Scale of measurement of the dependent variable (quantitative vs.qualitative)
C)Within or between subject design
D)The number of subjects used
E)All of these are correct

Question 17

A correlated t-test can be used to analyze the relationship between 2 variables except when_____.

A)the independent variable has 2 levels
B)the independent variable is between subject in nature
C)the groups have been matched on some variable
D)the dependent variable is quantitative in nature
E)except nothing,all of these are correct

Question 18

For an experiment that includes four levels of the single independent variable with 4 levels and,4 different groups are compared,the most appropriate statistical test is the:

A)correlate t-test
B)independent t-test
C)one-way ANOVA (between subjects)
D)ANOVA (repeated measures)
E)Two-way ANOVA (between subjects)

Question 19

If you set your alpha level at .01,nondirectional,and you are using a correlated t-test and have a total of 30 subjects what is your table or critical t ?

A)2.045
B)2.048
C)2.756
D)1.699
E)none of these

Question 20

The main difference between parametric & nonparametric statistics is that parametric statistics:

A)Makes no assumptions about the population distribution
B)Assumes that the population distribution is normally skewed
C)Assumes that the population distribution is normally distributed
D)Cannot be used when dependent measured on ratio & interval scales
E)All of these are correct

Question 21

The reason(s)why a One-way ANOVA is used to compare 3 or more means and not several t-tests is because:

A)It is too time consuming
B)It does not take into account variance
C)Increases the chance of a Type I error
D)Both a & b are correct
E)Both a & c are correct

Question 22

Multivariate Analysis of Variance (MANOVA)is used when there is_____.

A)Two levels of the independent variable
B)Two or more levels of the dependent variable
C)Nominal scales of measurement
D)None of these

Question 23

The F test for a one-way repeated measures ANOVA is appropriate when the dependent variable is_____in nature and measured on a level that approximates_____level of measurement.

A)quantitative;interval
B)quantitative;ratio
C)qualitative;nominal
D)qualitative;interval
E)both a & b

Question 24

When the independent variable is quantitative and the dependent variable is qualitative,an appropriate method of data analysis could be

A)Pearson correlation.
B)polychotomous logistic regression.
C)Kruskal-Wallis test.
D)none of these

Question 25

In the example above it was found that the higher your GPA is the higher you scored on the SAT which indicates_____.

A)positive correlation
B)negative correlation
C)a significant correlation
D)significant difference between groups
E)none of these

Question 26

In the example above if the researcher were able to reject the null hypothesis and state based on the color of the ice cream the ratings were different how would you determine the nature of the relationship?

A)Turkey HSD test
B)Inspection of the means
C)Square root of total variance
D)Eta-squared
E)None of these

Question 27

When both of the variables under study are qualitative and between-subjects in nature,the appropriate method of data analysis is the

A)independent groups t test.
B)one-way between-subjects analysis of variance.
C)two-way between-subjects analysis of variance.
D)chi square test.

Question 28

If group means are the focus of an analysis where the independent variable is qualitative and the dependent variable is quantitative,which of the following are further considerations which must be given before the type of test is chosen?

A)The number of values of the dependent variable.
B)The level of measurement of the dependent variable.
C)The extent of violation of distributional assumptions.
D)all of these

Question 29

A consumer psychologist was interested in the influence that the color of the ice cream has on perceived taste.Four hundred subjects were asked to taste 3 different types of vanilla ice cream.Each type was identical in ingredients except they were different shades of yellow.After tasting each ice cream,the subject was asked to rate the ice cream on a scale 1-5 (5 being more favorable: interval scale).What statistical test should be used?

A)Chi-Square
B)Correlated t-test
C)One-way ANOVA Repeated measures
D)One-way ANOVA between groups
E)None of these

Question 30

If you want to determine if GPA is correlated with SAT scores what test would you use?

A)correlate t-test
B)independent t-test
C)Pearson correlation
D)Spearman rank-order
E)both c & d

Question 31

Dr.James used a Pearson correlation and found that there was a relationship between the number of publications of faculty members and the number of graduate programs offered (r=.85).To determine the nature of the relationship what should Dr.James do?

A)Inspect the sign of r (positive or negative)
B)Use eta-squared
C)Perform a post-hoc test
D)Use coefficient of determination r²
E)both b & d

Question 32

A parametric test should NOT be used in situations where

A)the dependent variable is measure on an interval level.
B)the distributional assumptions of the parametric test have been violated but the test is robust to these violations.
C)the distributional assumptions of the parametric test have been violated and the test is not robust to these violations.
D)none of these

Question 33

When the independent variable is qualitative and the dependent variable is quantitative,most researchers will NOT use a parametric test in situations where the dependent measure has only two or three values.

Question 34

When the independent variable is qualitative and the dependent variable is quantitative,the choice of the appropriate method of data analysis depends on

A)whether a parametric or nonparametric test is desired.
B)whether the independent variable is between-subjects or within-subjects in nature.
C)the number of levels that characterize the independent variable.
D)all of these

Question 35

What statistical test would you use if your dependent variable is measured on an nominal scale and your independent variable is measured on a nominal scale and you are using a within subject design?

A)Wilcoxon Signed-Rank
B)Correlated t-test
C)One-way ANOVA Repeated measures
D)Chi-Square
E)None of these

Question 36

In a one-way analysis of variance (repeated measures)the strength of the relationship is determined by: _____.

A)eta-squared
B)Tukey HSD test
C)inspection of the means
D)square root of total variance
E)none of these

Question 37

Dr.Hinson used a One-Way ANOVA,repeated measures and has set his alpha level at .05 nondirectional.He reported the following: F(2,19)=6.99.Based on this information what can you conclude?

A)There was a significant difference between groups (reject null hypothesis)
B)There was no significant difference between groups (do not reject null hypothesis)
C)A type II error has been committed
D)There is a significant main effect of your independent variable
E)None of these

Question 38

A psychologist's is studying the affects of Ritalin on concentration in 40 children that have been matched on age and IQ.Twenty of the children receive Ritalin and are asked to take an exam that measures their concentration.The other 20 children did not receive Ritalin,but a placebo pill and took the same exam.After graphing this data,the researcher realizes that the data is not normally distributed.What should he do?

A)Increase your sample size
B)use a nonparametric test to analyze data
C)Give up and start over
D)Both a & b
E)None of these

Question 39

If one wishes to compare groups ranks,then a _____ method should be used.

A)parametric
B)nonparametric
C)either a parametric or a nonparametric
D)it depends on the number of levels of the independent variable

Question 40

A psychologist's is studying the affects of Ritalin on concentration in 40 children that have been matched on age and IQ.Twenty of the children receive Ritalin and are asked to take an exam that measures their concentration.The other 20 children did not receive Ritalin,but a placebo pill and took the same exam.What statistical test is appropriate for this study?

A)Correlate t-test
B)Independent t-test
C)One-way ANOVA (between subjects)
D)ANOVA (repeated measures)
E)Two-way ANOVA (between subjects)

Question 41

If ranks are the chosen focus of study,then _____ methods should be used.

A)experimental
B)multiple
C)parametric
D)nonparametric

Question 42

If the dependent variable is measured on an ordinal level that seriously departs from _____,then a nonparametric test might be pursued instead of a parametric test.

A)interval characteristics
B)normality
C)nominal characteristics
D)none of these

Question 43

Which of the following is used to examine relations among three or more qualitative variables simultaneously?

A)factor analysis
B)log-linear analysis
C)Bayesian procedure
D)multivariate analysis of variance

Question 44

The Hotelling T² test is used when a multivariate analysis of variance is applied to

A)a single independent variable that has only two levels
B)a single independent variable that has more than two levels
C)two or more independent variables,each of which has only two levels
D)two or more independent variables,each of which has more than two levels

Question 45

In situations where the dependent measure has only two or three values,most researchers will not use a _____ test.

A)chi square
B)log-linear
C)parametric
D)nonparametric

Question 46

In the equation Y =
,which of the following symbols is a regression coefficient?

A)
B)
₁
C)
₂
D)
E)both b and c

Question 47

A parametric procedure would not typically be used when a quantitative dependent variable has only two levels because the _____.

A)dependent variable can not be normally distributed
B)two levels might be correlated
C)data could not be compared
D)homogeneity of variance could not be determined

Question 48

If a dependent variable turns out to be highly skewed,then it might be decided to focus on a _____ rather than a _____ to analyze the relationship between two variables.

A)mean;median
B)median;mean
C)mean;mode
D)mode;Pearson correlation

Question 49

If means are the focus,then a _____ test is usually the analysis of choice.

A)chi square
B)log-linear
C)parametric
D)nonparametric

Question 50

When both of the variables under study are quantitative and between-subjects in nature,the appropriate method of data analysis is the

A)logistic regression.
B)Mann Whitney U test.
C)Spearman rank-order correlation.
D)none of these

Question 51

If both variables are qualitative,but one or both are within-subjects in nature,_____ of the statistical tests we have considered can be used.

A)all 12
B)several
C)only one
D)none

Question 52

A nonparametric test should be pursued if there is a reason to believe that the _____ of the corresponding parametric test have been violated to the extent that the parametric test is not robust.

A)kurtotic assumptions
B)theoretical assumptions
C)population parameters
D)distributional assumptions

Question 53

Which of the following is not an important question when analyzing bivariate relationship?

A)Is there a relationship between the variables?
B)What is the nature of the relationship?
C)How strong is the relationship?
D)a,b,and c are important.

Question 54

Which of the following is a first step in determining the appropriate bivariate test?

A)Determine the number of subjects needed for a given level of power.
B)Identify the independent and dependent variables as qualitative or quantitative.
C)Determine the appropriate alpha level.
D)Select the appropriate post hoc test.

Question 55

The major factor to be weighed when deciding whether to use a parametric or a nonparametric test is the decision about what aspect of the _____ you want to compare groups on.

A)independent variable
B)distribution
C)sample

Question 56

Multivariate statistics involve analyzing the relationship between

A)three or more variables.
B)two independent variables.
C)one independent variable and one dependent variable.
D)none of these

Question 57

Multiple regression involves two or more

A)criterion variables.
B)predictor variables.
C)dependent variables.

Question 58

A researcher is interested in whether the number of years working in a company (1,5,or 10)might have an effect on job satisfaction (rating scale 1-50),employer satisfaction (rating scale 1-20)and perception of job difficulty (rating scale 1-10).Which of the following tests would be used to analyze this data?

A)Two-way between-subjects analysis of variance
B)Two-way repeated measures analysis of variance
C)Two-way between-within analysis of variance
D)Multivariate analysis of variance

Question 59

A researcher is interested in the number of errors in the following conditions.Thirty people are given a list of word pairs to study;half of the pairs are semantically related and half are unrelated.Another thirty people are given a different list of word pairs to study;half of the pairs are orthographically related and half are unrelated.Later all 60 people are asked to recall the word pairs.Which of the following tests would be used to analyze this data?

A)Two-way between-subjects analysis of variance
B)Two-way repeated measures analysis of variance
C)Two-way between-within analysis of variance
D)Multivariate analysis of variance

Question 60

When both of the variables under study are qualitative and between-subjects in nature,the appropriate method of data analysis is the:

A)chi square test
B)Pearson correlation
C)multiple regression
D)none of these

Question 61

To determine if the correlations among a set of variables can be accounted for by a one or more underlying factors,a _____ should be used.

A)log-linear analysis
B)factor analysis
C)multiple regression
D)Spearman correlation

Question 62

The joint influence of two within-subjects qualitative independent variables can be studied using:

A)correlated groups t test
B)two-way repeated measures analysis of variance
C)Hotelling T² test
D)factor analysis

Question 63

An index of the strength of the relationship between the criterion variable and the set of predictor variables is provided by the _____,symbolized by _____.

A)squared t;t²
B)squared multiple correlation coefficient;R²
C)regression coefficient;á
D)multiple r;r²

Question 64

Statistical techniques for analyzing the relationship between three or more variables are referred to as:

A)nonparametric statistics
B)trivariate statistics
C)multiple correlations
D)multivariate statistics

Question 65

Alternatives to the "null hypothesis testing approach" include:

A)estimating effect sizes
B)Bayesian statistics
C)a and b
D)none of these

Question 66

The strength of the relationship is usually indexed by the 95% confidence interval.

Question 67

When applied to a single independent variable having only two levels,multivariate analysis of variance is referred to as:

A)log-linear analysis
B)factor analysis
C)Hotelling T² test
D)multiple regression

Question 68

A _____ is often used to test whether the groups have different population means on the dependent variables considered jointly.

A)multivariate analysis of variance
B)multiple regression
C)Hotelling T² test
D)factor analysis

Question 69

_____ will usually be applied to research questions requiring that three or more qualitative variables be simultaneously examined.

A)Factor analysis
B)Multivariate analysis of variance
C)Log-linear analysis
D)Multiple regression

Question 70

If the dependent variable is measured on an ordinal level that seriously departs from interval characteristics,then a _____ test is typically used.

A)parametric
B)nonparametric

Question 71

______ is the statistic used to indicate the strength of relationship between a criterion variable and several predictors in the context of a linear model

A)Eta²
B)R²
C)r²

Question 72

Statistical analyses that are designed to analyze the relationship between three or more variables simultaneously are called _____ tests.

A)interaction comparison
B)simple main effects
C)multivariate
D)correlational

Question 73

Given two between-subjects variables that are multi-valued and measured on a level that at least approximates interval characteristics,the most common method of analysis is:

A)Spearman rank-order correlation
B)Pearson correlation
C)point-biserial correlation
D)nonparametric correlation

Question 74

Inference of a relationship between two variables depends on the comparison of means.

Question 75

The appropriate means of analysis for research designs involving one between-subjects qualitative independent variable and one within-subjects qualitative independent variable would be a:

A)log-linear analysis
B)two-way between-within analysis of variance
C)Hotelling T² test
D)multiple regression

Question 76

Under the correct circumstances a nonparametric tests can be as powerful as a parametric test.

Question 77

Some psychologists have argued that the question "is there a relationship between two variables" is meaningless,in part because:

A)psychological research is difficult to carry out
B)the alternative hypothesis is almost always false
C)the null hypothesis is almost always false
D)empirical relationships are,in fact,very rare

Question 78

The goal of _____ is to determine if the correlations among a set of variables can be accounted for by one or more underlying dimensions,or _____.

A)factor analysis;factors
B)regression analysis;regressions
C)log-linear analysis;log odds
D)none of these

Question 79

Multiple regression analysis involves the prediction of a _____ variable from two or more _____ variables.

A)nominal;ordinal
B)within-subjects;between-subjects
C)predictor;criterion
D)criterion;predictor

Question 80

In multiple regression analysis,each predictor has a "slope" (â)that is called a:

A)factor
B)squared multiple correlation coefficient
C)regression coefficient
D)multiple r