Deck 7: Examining Differences Between Means: The T-Test

Question 1

If an experimenter conducts a t-test to see whether the responses of subjects in a control group differ from those of an experimental group, which of the following outcomes will yield the highest t-value?

A)If there are 52 subjects in each condition, the mean difference between the responses of the control group and the experimental group is 10 and both have standard deviations of 3.5.
B)If there are 52 subjects in each condition, the mean difference between the responses of the control group and the experimental group is 20 and both have standard deviations of 3.5.
C)If there are 52 subjects in each condition, the mean difference between the responses of the control group and the experimental group is 20 and both have standard deviations of 5.2.
D)If there are 15 subjects in each condition, the mean difference between the responses of the control group and the experimental group is 20 and both have standard deviations of 3.5.
E)If there are 15 subjects in each condition, the mean difference between the responses of the control group and the experimental group is 20 and both have standard deviations of 5.2.

Answer 1

If there are 52 subjects in each condition, the mean difference between the responses of the control group and the experimental group is 20 and both have standard deviations of 3.5.

Question 2

A clinical neuropsychologist develops a new treatment for recovery from head trauma and administers it to one half (randomly selected)of a sample of 50 people.He measures the level of cognitive functioning one month after the trauma and analyses the data in SPSS.The output of a between-participant t-test is shown below: Which of the following statements is true?

A)There are good grounds to suspect that the variances are unequal.
B)We should reject the results of this test due to the unequal variances.
C)Both (a)and (b).
D)It is not surprising that the cognitive functioning is variable following head trauma so we should pay no attention to unequal variances, especially as the t-test is robust to violation of assumptions.
E)All of the above.

Answer 2

There are good grounds to suspect that the variances are unequal.

Question 3

A psychology professor has set a test for 50 students with each question passed scored as 1 and each question failed scored as 0 from a total of 100 questions.One week later the professor then asked the students to write down all the things they had passed that they could remember.These were scored as correctly remembered or incorrectly remembered by coders who were blind to the experimental conditions.He then conducts a within-samples t-test to compare the number passed with the number remembered and provides the SPSS output below. Which of the following statements is true?

A)The correlation between the passed and remembered things is too small to be useful
B)The appropriate terminology in psychology is to refer to the blind coders as "visually impaired"
C)The mean for remembered is too small to be valid or reliable
D)The big differences in standard deviations between the variables is worth examining.
E)A between-samples t-test should have been used.

Answer 3

The big differences in standard deviations between the variables is worth examining.

Question 4

After careful ethical review of planned procedures, a researcher tests a new brain activation drug by taking a sample of 50 volunteers and randomly assigning 25 participants to the treatment condition and 25 to the control (placebo)condition.She then carries out an fMRI scan to assess the level of activation in regions of the participants' brains 30 minutes after administering the drug.She carries out a between-participants t-test in SPSS which provides the output shown below: What further advice is appropriate to give the researcher here?

A)The variances look as if they could be unequal.
B)It is worth ensuring that the distributions of the scores are normal.
C)The negative confidence interval suggests she needs to reduce statistical uncertainty further.
D)The t-test should be accompanied with a measure of effect size such as Cohen's d.
E)Both (a)and (d).

Question 5

Which of the following does not compromise the results of t-tests?

A)Evidence that scores in the two conditions being compared have different variance.
B)Non-independence of observations in the case of between-subjects tests.
C)Evidence that scores on a variable are uniformly distributed.
D)Evidence that scores on a variable are not normally distributed.
E)Suggestions that the measures on which scores are obtained do not measure what they are supposed to.

Question 6

If an experimenter conducts a t-test to see whether the responses of subjects in a control group differ from those of an experimental group, which of the following factors will not affect the significance of that test?

A)The alpha level chosen by the experimenter
B)The difference between the sample means
C)The sample variances
D)The size of the samples
E)The distribution of the samples

Question 7

If an experimenter conducts a t-test to see whether the responses of subjects in a control group differ from those of an experimental group, which of the following outcomes will yield the highest t-value?

A)If there are 30 subjects in each condition and mean difference between the responses of the control group and the experimental group is 2 and both have standard deviations of 2.
B)If there are 40 subjects in each condition and mean difference between the responses of the control group and the experimental group is 1 and both have standard deviations of 1.
C)If there are 40 subjects in each condition and mean difference between the responses of the control group and the experimental group is 2 and both have standard deviations of 1.
D)If there are 40 subjects in each condition and mean difference between the responses of the control group and the experimental group is 1 and both have standard deviations of 2.
E)If there are 40 subjects in each condition and mean difference between the responses of the control group and the experimental group is 2 and both have standard deviations of 2.

Question 8

A clinical neuropsychologist develops a new treatment for recovery from head trauma and administers it to one half (randomly selected)of a sample of 50 people.He measures the level of cognitive functioning one month after the trauma and analyses the data in SPSS.The output of a between-participant t-test is shown below: Which of the following statements is true?

A)A measure of effect size called d can be constructed using t and the df.
B)A measure of effect size called r can be constructed using t and the df.
C)The effect here will be large because the effect is highly significant with F = 11.364 and p = .001.
D)The effect here will be large because the effect is highly significant with t (48)= 5.744 and p < .001.
E)The effect here will be large because the effect is highly significant with t (37.363)= 5.744 and p < .001.

Question 9

"The hypothetical distribution of differences between the means of two random samples drawn from the same population.It has a mean of 0 and a standard deviation that decreases with the square root of the sample size." What is this a glossary definition of?

A)The hypothetical distribution of means.
B)The sampling distribution of the difference between means.
C)The random distribution of scores.
D)The t-distribution.
E)The population distribution of difference between means.

Question 10

After careful ethical review of planned procedures, a researcher tests a new brain activation drug by taking a sample of 50 volunteers and randomly assigning 25 participants to the treatment condition and 25 to the control (placebo)condition.She then carries out an fMRI scan to assess the level of activation in regions of the participants' brains 30 minutes after administering the drug.She carries out a between-participants t-test in SPSS which provides the output shown below: Which of the following statements is true?

A)The assumption of equal variance is irrelevant.
B)The assumption of equal variance is violated.
C)The assumption of equal variance might be violated but we cannot tell from this output as we only have standard deviations
D)We cannot judge whether the assumption of equal variances is violated as we need to see the population distribution.
E)None of the above.

Question 11

A clinical neuropsychologist develops a new treatment for recovery from head trauma and administers it to one half (randomly selected)of a sample of 50 people.He measures the level of cognitive functioning one month after the trauma and analyses the data in SPSS.The output of a between-participant t-test is shown below: Which of the following statements is true?

A)The test shows the treatment works.
B)The test shows the treatment works so long as we can rule out problems with internal validity.
C)The test shows the treatment works so long as we can rule out problems with internal validity and external validity (e.g., selection X treatment effects).
D)The test shows that the effects of the treatment are not due to chance.
E)The test provides us with good grounds to be confident that the observed difference is not due to chance.

Question 12

A psychology professor has set a test for 50 students with each question passed scored as 1 and each question failed scored as 0 from a total of 100 questions.One week later the professor then asked the students to write down all the things they had passed that they could remember.These were scored as correctly remembered or incorrectly remembered by coders who were blind to the experimental conditions.He then conducts a within-samples t-test to compare the number passed with the number remembered and provides the SPSS output below. Which of the following statements is false?

A)The significant difference shows that the time gap between the test is too large to be reliable.
B)The p-value is too small to be useful.
C)The scores cannot be compared by t-test because one measure recognition and the other recall.
D)A manipulation check would be useful for the remembrance measure.
E)All of the above.

Question 13

After careful ethical review of planned procedures, a researcher tests a new brain activation drug by taking a sample of 50 volunteers and randomly assigning 25 participants to the treatment condition and 25 to the control (placebo)condition.She then carries out an fMRI scan to assess the level of activation in regions of the participants' brains 30 minutes after administering the drug.She carries out a between-participants t-test in SPSS which provides the output shown below: Which of the following statements is false?

A)An F value of .000 is significant with an alpha level of .001.
B)The probability is .036 that the difference between conditions is due to chance.
C)The probability is .036 that the differences between conditions could be produced by a random process of selecting two groups of 25 people from that population.
D)Both (a)and (b).
E)All of the above.

Question 14

A psychology professor has set a test for 50 students with each question passed scored as 1 and each question failed scored as 0 from a total of 100 questions.The professor conducted a one sample t-test shown in the SPSS output below. What are the values being compared in the test?

A)A test value of 50 chosen by the professor and the mean performance of the class.
B)The N of 50 and the mean performance of the class.
C)The df (49)and the upper bound of the 95% confidence interval.
D)A test value of 50 and/or the N of 50 with the mean performance of the class.
E)The rounded variance (standard deviation squared = 57.76)minus the standard deviation (7.6)= 50 (rounded)and the mean performance of the class.

Question 15

"An estimated range of values for a population parameter.The size of this is expressed as a percentage".What is this a glossary definition of?

A)Estimated variance.
B)Population variance.
C)Confidence interval.
D)Population interval.
E)Estimated confidence.

Question 16

"The hypothesis that the research reveals no effect (e.g.that two samples are drawn from the same population and that there is no informative difference between their means).Also referred to as H₀.This is the hypothesis that statistical procedures allow researchers to reject." What type of hypothesis is this a glossary definition of?

A)The experimental hyptothesis
B)The alternative hyptothesis
C)The null hypothesis
D)The first hypothesis
E)The second hypothesis

Question 17

"The largest probability that the researcher is prepared to accept in order to reject the null hypothesis.The probability of a random process producing a result as big as the one obtained has to be smaller than this for a statistical test to be significant.It is equal to the probability of making a Type I error and should be set before research is carried out." What is this a glossary definition of?

A)Alpha level
B)Beta level
C)Gamma level
D)Delta level
E)None of the above.

Question 18

"The hypothetical distribution of expected differences between the means of randomly chosen samples drawn from the same population divided by an estimate of the standard error." What construct is this a glossary definition of?

A)The distribution of the difference between means
B)The sampling distribution of the difference between means
C)The t-distribution
D)The x-distribution
E)The z-distribution

Question 19

After careful ethical review of planned procedures, a researcher tests a new brain activation drug by taking a sample of 50 volunteers and randomly assigning 25 participants to the treatment condition and 25 to the control (placebo)condition.She then carries out an fMRI scan to assess the level of activation in regions of the participants' brains 30 minutes after administering the drug.She carries out a between-participants t-test in SPSS which provides the output shown below: What is the best summary of the outcome of this t-test?

A)The difference between conditions is significant.
B)The differences between conditions is significant with an alpha level of .05.
C)The difference between conditions is not significant.
D)The differences between conditions is marginally significant.
E)The treatment is obviously effective.

Question 20

After careful ethical review of planned procedures, a researcher tests a new brain activation drug by taking a sample of 50 volunteers and randomly assigning 25 participants to the treatment condition and 25 to the control (placebo)condition.She then carries out an fMRI scan to assess the level of activation in regions of the participants' brains 30 minutes after administering the drug.She carries out a between-participants t-test in SPSS which provides the output shown below: Which of the following statements is true?

A)The result is not significant because the p value is .985.
B)The result is not significant because a negative t-test indicates a negative result.
C)There are 48 degree of freedom because N is 50 and there are two groups.
D)The scan measures processes within-participants' brains so the researcher should have performed a within-participants t-test.
E)We need to make sure the correlation between the control and treatment scores is more than .3 before we can confidently use the t-test.

Question 21

If a researcher is concened that the relationship between two variables is non-linear, which of the following might be good advice to help them improve their research?

A)Consider values of variables that have not been sampled and considering whether the study is compromised by failing to include these.
B)Draw a scatterplot and see if data fall on a straight line.If they do not, be alert to the limitations of tests for linear relationships.
C)Try to understand what is constraining responses on the affected variables.If possible, suggest alternative measures.
D)Suggest alternative interpretations of the correlation.Consider conducting experimental studies that might address issues of causality.

Question 22

A personality researcher revisits the hypothesis that frustration leads to regression.She collects data from 49 female Internet users measuring their sense of personal frustration and their self-ratings of aggression.A scatterplot is shown below. Which of the following statements is true?

A)The research shows that frustration causes aggression.
B)The research shows that aggression causes frustration.
C)The research shows that either frustration causes aggression or aggression causes frustration but we need an experiment to decide which one is true.
D)It is impossible to determine the nature of the causal relationship between variables in this type of research.
E)It looks like frustration causes aggression but we need to know whether the effect is significant before we can draw this conclusion.

Question 23

A personality researcher revisits the hypothesis that frustration leads to regression.She collects data from 49 female Internet users measuring their sense of personal frustration and their self-ratings of aggression.A scatterplot is shown below. Which of the following statements is true?

A)The research is not valid as it only uses females.
B)The research is not valid as the sample size is too small.
C)The research is valid but it is not reliable as all the participants were female.
D)The research is internally valid but it is not externally valid as all the participants were female.
E)None of the above.

Question 24

"A test with two rejection regions that is based on the alternative hypothesis that research reveals an effect (e.g.that two means are different).The nature of any departure from randomness (e.g.which mean will be larger)is not specified." What is this a glossary definition of?

A)A one-tailed test.
B)A two-tailed test.
C)Two one-tailed tests.
D)A bimodal test.
E)A binomial test.

Question 25

"The conditions that must be satisfied before we can use a statistical test with confidence.Those of the t-test are that the samples are independent and randomly sampled from a normally distributed population that has the same variance for all samples".What is this a glossary definition of?

A)Caveats.
B)Assumptions.
C)Operational procedures.
D)Standard operating procedures.
E)Statistical protocols.

Question 26

"A common statistical purpose or goal of correlational analyses.The aim is to be able to estimate (with some uncertainty)a person's score on one variable by knowing that person's score on another variable (or variables)." What is this a glossary definition of?

A)Correlation.
B)Prediction.
C)Estimation.
D)Validation.
E)Explanation.

Question 27

A personality researcher revisits the hypothesis that frustration leads to regression.She collects data from 49 female Internet users measuring their sense of personal frustration and their self-ratings of aggression (on a scale from 1 to 7).A scatterplot is shown below. Which of the following statements is true?

A)The scatterplot suggests there is a positive correlation between frustration and aggression.
B)The scatterplot suggests there is a negative correlation between frustration and aggression.
C)The scatterplot suggests there is no correlation between frustration and aggression.
D)The scatterplot provides no information about correlations.
E)The scatterplat suggests that these data need to be subjected to Analysis of Variance (ANOVA).

Question 28

"The hypothesis that the research reveals an effect (e.g.that two samples are drawn from different populations and that there is an informative difference between their means).Also referred to as H₁.This is normally the hypothesis that the research was designed to test." What type of hypothesis is this a glossary definition of?

A)The experimental hyptothesis
B)The alternative hyptothesis
C)The null hypothesis
D)The first hypothesis
E)The second hypothesis

Question 29

"The ability of a statistical test to reveal effects that exist in the data.Increasing this reduces inferential uncertainty.Formally, it is equal to 1 − β, where β is the probability of a Type II error." What construct is this a glossary definition of?

A)Type I error
B)The probability of Type I error
C)Power
D)Statistical significance
E)Statistical confidence

Question 30

"Involving the relationship between two variables".What is this a glossary definition of?

A)Bivariate.
B)Correlational.
C)Dualistic.
D)Birelational.
E)Bidirectional.

Question 31

A personality researcher revisits the hypothesis that frustration leads to regression.She collects data from 49 female Internet users measuring their sense of personal frustration and their self-ratings of aggression.A scatterplot is shown below. Which of the following statements is true?

A)The scatterplot suggests that there are some potential outliers.
B)There are no obvious outliers in the scatterplot.
C)Outlying scores should have been discarded before they were reported.
D)Outliers are not a concern in correlational analysis.
E)Both (b)and (c).

Question 32

"The mistaken belief that a statistically significant result in a psychological study constitutes a psychologically significant finding".What fallacy is this a glossary definition of?

A)The significance fallacy.
B)The causal fallacy.
C)The operational fallacy.
D)The psychological fallacy.
E)The gambler's fallacy.

Question 33

An educational psychologist administered a number of ability measures to 120 14-year olds.The measures included an IQ test, and tests of mathematics, spelling, and general verbal ability as well as a test of the amount of time it took the students to learn material from the new curriculum.The correlation table is shown below: Based on the output which of the following statements is false?

A)Learning time is an unreliable measure.
B)By Cohen's standards spelling ability is strongly related to verbal ability.
C)Scores on all of the measures are caused by IQ.
D)The lower correlations for learning time from left to right across the table reflect the effects of truncated range.
E)All of the above.

Question 34

A personality researcher revisits the hypothesis that frustration leads to regression.She collects data from 49 female Internet users measuring their sense of personal frustration and their self-ratings of aggression.A scatterplot is shown below.A colleague who comments on the research suggests that frustration and aggression are both caused by early childhood experiences. Which of the following statements is true?

A)The colleague is suggesting this is an example of a voodoo correlation.
B)The colleague's alternative hypothesis appears to be Freudian and so can be ignored in modern psychology.
C)The colleague's alternative hypothesis sounds Freudian and while this is relevant for modern psychology it has nothing to do with online behavior.
D)The colleague is proposing that the correlation between frustration and aggression is spurious.
E)Both (a)and (c).

Question 35

"A relationship between variables where high scores on the first variable are associated with low scores on the second variable, and low scores on the first are associated with high scores on the second." What type of relationship is this a glossary definition of?

A)An indeterminate correlation.
B)A negative correlation.
C)A positive correlation.
D)A high correlation.
E)A low correlation.

Question 36

"The degree to which scores above the mean on one variable are associated with scores above the mean on another variable.What is this a glossary definition of?

A)Covariance
B)Pearson's product-moment correlation coefficient
C)A covariance coefficient.
D)A product-moment coefficient.
E)A linear coefficient.

Question 37

"Selecting part of the sample so that there is not a full range of scores on a variable being measured".What is this a glossary definition of?

A)Selection bias.
B)Experimenter bias.
C)Contractional correlation.
D)Restricted range.
E)None of the above.

Question 38

An educational psychologist administered a number of ability measures to 120 14-year olds.The measures included an IQ test, and tests of mathematics, spelling, and general verbal ability as well as a test of the amount of time it took the students to learn material from the new curriculum.The correlation table is shown below: The research is criticized by other psychologists.On the basis of the information provided which of the following is a potentially valid criticism?

A)Correlation is not causation so the researcher should have experimentally manipulated IQ.
B)The research involves a within-participants design, so in order for the test to be valid the correlation needs to be greater than .3.
C)There is more than one variable so this is a case of multiple regression to the mean.
D)There is no independent variable so it is not an experiment so the results can tell us nothing about psychology.
E)None of the above.

Question 39

A personality researcher revisits the hypothesis that frustration leads to regression.She collects data from 49 female Internet users measuring their sense of personal frustration and their self-ratings of aggression (on a scale from 1 to 7).A scatterplot is shown below. Which of the following statements is true?

A)The scatterplot suggests that Aggression scores are positively skewed.
B)The scatterplot suggests that Aggression scores are negatively skewed.
C)The scatterplot provides no information about skew.
D)Scatterplots should not be used in correlational analysis as this introduces experimenter bias.
E)Scatterplots can be used in correlational analysis but they are a major source of demand characteristics.

Question 40

A personality researcher revisits the hypothesis that frustration leads to regression.She collects data from 49 female Internet users measuring their sense of personal frustration and their self-ratings of aggression.A scatterplot is shown below. Which of the following statements is true?

A)This is a classic example of a curvilinear relationship.
B)This is a classic example of truncated range.
C)The variables appear to be moderately to highly correlated.
D)This is a classic example of a selection by interaction effect.
E)Both (c)and (d).

Question 41

"Selecting part of the sample so that there is not a full range of scores on a variable being measured." What is this a glossary definition of?

A)Restricted range.
B)Maximum value fallacy.
C)Floor effect.
D)Ceiling effect.
E)Regression to the mean.

Question 42

A developmental psychologist developed a training program to encourage children to reduce their Internet usage.To examine its efficacy he recruited a sample of 120 children - of whom 60 were aged 11 years and 60 were aged 13.Half of the participants were randomly assigned to the training program and the amount of online usage was measured after one month.Findings were subjected to two-way ANOVA, from which the output was as follows: Which of the following statements is true?

A)The large number of degrees of freedom associated with the error term is a cause for concern.
B)The large number of degrees of freedom associated with the error term is only a cause for concern is the assumption of homogeneity of regression to the mean is violated.
C)The assumption of independence is irrelevant here because the researcher used randomization.
D)The effect for condition is of medium size by Cohen's standards.
E)The probability of randomly sampling a difference of this size is .001.

Question 43

Which of the followings statements is true about Analysis of Variance (ANOVA)?

A)If sample variances are equal you should seriously consider using a non-parametric test.
B)If sample variances are equal you should seriously consider using analysis of covariance (ANCOVA).
C)If sample variances are unequal (e.g., one is more than four times bigger than the other)you should seriously consider using a non-parametric test.
D)If sample variances are unequal (e.g., one is more than four times bigger than the other)you should seriously consider using analysis of covariance (ANCOVA).
E)None of the above.

Question 44

"A summary table whose rows relate to the various sources of variation and columns relate to the corresponding sums of squares, degrees of freedom, and mean squares, as well as values of F, p and R²." What is this a glossary definition of?

A)An F-table.
B)A covariance matrix.
C)A variation table.
D)An ANOVA table.
E)none of the above.

Question 45

A developmental psychologist developed a training program to encourage children to reduce their Internet usage.To examine its efficacy he recruited a sample of 120 children - of whom 60 were aged 11 years and 60 were aged 13.Half of the participants were randomly assigned to the training program and the amount of online usage was measured after one month.Findings were subjected to two-way ANOVA, from which the output was as follows: Which of the following statements is true?

A)The effect for age is large because it is less than .05
B)The effect sizes cannot be interpreted in this output due to overlap between variables
C)The results need to be interpreted with some caution as the table shows there were 179.414 errors in the output
D)The results cannot be compared to a random process because the study involved randomization (which is quite different to random sampling)
E)None of the above.

Question 46

"An estimate of variance between cells that provides the numerator of an F-ratio. " Which construct is this a glossary definition of?

A)Within-cells mean square (MS_W).
B)Between-cells mean square (MS_B).
C)Grand mean.
D)Mean squares (MS).
E)Grand mean sqares (_gMS).

Question 47

How might one deal with the possibility of outliers in Analysis of Variance (ANOVA)?

A)Examine the distribution of scores to see if it contains outliers.If it does, remove them.
B)Examine the distribution of scores to see if it contains outliers.If it does, see if this can be corrected by an appropriate statistical transformation of the sample data.If this is not possible, consider removing individual outliers.Ensure you report any transformations that you make.
C)Examine the distribution of scores to see if it contains outliers.If it does, see if this can be corrected by an appropriate statistical transformation of the sample data.If this is not possible, consider recoding outliers to the same value as the nearest nonoutlier.Ensure you report any transformations that you make.
D)Both (b)and (c).
E)Do not worry: Analysis of Variance automatically corrects for outliers.

Question 48

A developmental psychologist developed a training program to encourage children to reduce their Internet usage.To examine its efficacy he recruited a sample of 120 children - of whom 60 were aged 11 years and 60 were aged 13.Half of the participants were randomly assigned to the training program and the amount of online usage was measured after one month.Findings were subjected to two-way ANOVA, from which the output was as follows: Which of the following is true about the design of this study?

A)There are three factors: condition, age, and group
B)There are three factors: condition, age, and interaction.
C)There is one factor with two sub-factors: training program and age group.
D)There are two factors: condition and age group.
E)There is one factor - condition - because you ignore nonsignificant factors unless there is a confound.

Question 49

"A barrier to the correct interpretation of results that arises when the range of scores is limited because they are clustered in the region of the minimum scale value." What type of effect is this a glossary definition of?

A)Restricted range effect.
B)Limited range effect
C)Minimum value effect.
D)Floor effect.
E)Ceiling effect.

Question 50

A developmental psychologist developed a training program to encourage children to reduce their Internet usage.To examine its efficacy he recruited a sample of 120 children - of whom 60 were aged 11 years and 60 were aged 13.Half of the participants were randomly assigned to the training program and the amount of online usage was measured after one month.Findings were subjected to two-way ANOVA, from which the output was as follows: Which of the following statements is true?

A)The nonsignificant main effect for age group tells us we can ignore age.
B)The nonsignificant main effect for age group tells us we can ignore age so long as our alpha level is not significant.
C)The nonsignificant main effect for age group tells us we can ignore age so long as our alpha level is less than .225.
D)Despite the nonsignificant main effect for Age this IV is still relevant because it is implicated in a significant interaction effect.
E)Age is relevant because maturation effects may lead to central limits on randomization.

Question 51

A developmental psychologist developed a training program to encourage children to reduce their Internet usage.To examine its efficacy he recruited a sample of 120 children - of whom 60 were aged 11 years and 60 were aged 13.Half of the participants were randomly assigned to the training program and the amount of online usage was measured after one month.Findings were subjected to two-way ANOVA, from which the output was as follows: Which of the following statements is false?

A)The assumption of equal variances should be tested for a Hawthorne effect.
B)The interaction effect is significant with p = .003.
C)The effect of intercept can be removed by truncating the range.
D)The results cannot be interpreted because the corrected total is much smaller than the total.
E)All of the above.

Question 52

"Comparisons between cells that a researcher has planned to perform prior to a study being conducted.These constitute a subset of all possible comparisons and have usually been identified as important on the basis of a particular theory." What is this a glossary definition of?

A)Restricted comparisons.
B)Proper comparisons.
C)Post hoc comparisons.
D)Ad hoc comparisons.
E)A priori comparisons.

Question 53

Some social and political psychologists identified groups of people who were alarmed, concerned or cautious about climate change.They then measured the commitment of those people to take action on climate change and conducted a one-way ANOVA.The results of this are below: Which of the following statements is true?

A)The variances are dangerously unequal.
B)The variances are significantly unequal (Levene statistic < .05).
C)Analysis of variance is a special form of factor analysis and there is one significant factor here.
D)There are 2 degrees of freedom because there are three groups.
E)None of the above.

Question 54

In the figure below, what is the main clue to the fact that there may be a statistical interaction between the two independent variables (size and symmetry)?

A)The fact that one line is higher than the other.
B)The fact that both values are lower on the left than on the right.
C)The fact that the lines are not parallel.
D)The fact that mean reaction time is always between 120 and 200 ms.
E)The fact that the researcher has displayed the data using a line graph.

Question 55

"Experimental designs in which all levels of each factor occur within each level of the other factor.If an experiment involves the factors age (with two levels: young and old)and sex (with two levels: male and female), the old group will contain males and females, the young group will contain males and females, the male group will contain young and old, and the female group will contain young and old." What type of design is this a glossary definition of?

A)Complete design.
B)Multi-factor design.
C)Fully factorial design.
D)Demographic design.
E)Multivariate design.

Question 56

"The hypothetical distribution of the expected ratio of variability between group means divided by variability within group means." What distribution is this a glossary definition of?

A)F-distribution.
B)t-distribution.
C)z-distribution.
D)expected distribution.
E)variability distribution.

Question 57

A developmental psychologist developed a training program to encourage children to reduce their Internet usage.To examine its efficacy he recruited a sample of 120 children - of whom 60 were aged 11 years and 60 were aged 13.Half of the participants were randomly assigned to the training program and the amount of online usage was measured after one month.Findings were subjected to two-way ANOVA, from which the output was as follows: Which of the following is a fair summary of the results?

A)There is a main effect for condition.
B)There is a minor effect for condition.
C)There is a partial effect for condition.
D)There is a main effect for condition and a minor effect for training program.
E)There is an interaction effect for error.

Question 58

Some social and political psychologists identified groups of people who were alarmed, concerned or cautious about climate change.They then measured the commitment of those people to take action on climate change and conducted a one-way ANOVA.The results of this are below: Which of the following statements is true?

A)We need a two-way ANOVA to look at interactions between the groups.
B)We need to know the correlation between Alarmed and Cautious groups.
C)Judging from the means, ceiling and floor effects do not appear likely to be a problem.
D)The Alarmed group result is not valid it is a combination of regression to the mean and history effects.
E)All of the above.

Question 59

"The sum of the squared difference between all cell means and the grand mean multiplied by the cell size." Which construct is this a glossary definition of?

A)Sum of squares between cells (SS_B)
B)Within-cells mean square (MS_W).
C)Between-cells mean square (MS_B).
D)Grand mean.
E)Sum of mean squares (MS).

Question 60

"A pattern of two-way interaction, in which, when a line graph is drawn, the lines connecting the means for different levels of the same factor cross over each other.This pattern indicates that at one level of one of the factors (A)the mean is higher for one level of the other factor (X will be greater than Y ), but that at another level (B)this pattern is reversed (Y will be greater than X)." What is this a glossary definition of?

A)A complex interaction.
B)A cross-over interaction.
C)An AX-BY interaction.
D)An AB-XY interaction.
E)None of the above.

Question 61

A researcher, Bill, adopts a qualitative research strategy in which they want to focus on the meaning that computer-generated marketing letters have for those who receive them.To do this, they interview 16 people who have received such letters and subect their responses to thematic analysis.Which of the following is clearly true?

A)Bill is taking a hermeneutic approach.
B)Bill is committed to the principle of participant involvement.
C)Bill is committed to the principle of user involvement.
D)Bill is committed to the contingent repertoire.
E)Bill is committed to the empiricist repertoire.

Question 62

A clinical psychologist recruits a sample of 144 people experiencing anxiety disorders to test the effectiveness of a new therapy.However, 12 of the participants withdraw before the end of the program.Of the remaining 132 participants, 59 were randomly assigned to a control condition in which they received conventional treatment (shown as "Cond.1" below)and 25 of them were assessed to have reduced symptoms after treatment.At the same time, 69 participants were randomly assigned to a condition in which they received the new treatment (shown as "Cond.2")and, of these, 41 were assessed to have reduced symptoms after treatment.The output from a chi-square test (conducted using Preacher's, 2001, online tool)is shown below: Which of the following statements is true?

A)The unequal sample size is no problem for deciding whether the treatment is effective, the calculation takes account of this.
B)The unequal sample size is a concern because it can be a threat to internal validity in the form of a selection by maturation interaction.
C)The unequal sample size may be a concern because of a threat to internal validity in the form of differential mortality (e.g., people dropped out of the control condition because they were not receiving treatment).
D)It does not matter whether the research finding is valid because the results were not significant.
E)It does not matter whether the research finding is valid because the results were marginally significant.

Question 63

Which of the following statements is false?

A)Qualitative data can be structured or unstructured.
B)Qualitative data can be gathered for research purposes or can exist independently of research.
C)Qualitative researchers can be internal or external to their data.
D)The more structured qualitative data are, the more likely they are to be handled in widely agreed ways after they have been collected.
E)None of the above (i.e., all are true).

Question 64

"Events that are the combination of two or more categories of event.For example, where being a woman with an eating disorder is the consequence both of being a woman (vs.a man)and of having (vs.not having)an eating disorder." What is this a glossary definition of?

A)Combinational probability
B)Conditional probability.
C)Contingencies.
D)Degrees of freedom.
E)Freedom of degrees.

Question 65

Which of the following features do not routinely distinguish between quantitative methods and classical qualitative methods.

A)The form of the data.
B)The method of dtaa collection.
C)The form of data analysis.
D)The feel of data analysis.
E)The need for data analysis.

Question 66

To the extent that qualitative methods serve to emphasize the subjective, human aspects of the research process, which of the following statements is true?

A)They are compatible with what Gilbert and Mulkay (1984)refer to as the empiricist repertoire in scientific research.
B)They are compatible with what Gilbert and Mulkay (1984)refer to as the contingent repertoire in scientific research.
C)They are compatible with what Gilbert and Mulkay (1984)refer to as the hermeneutic repertoire in scientific research.
D)Both (a)and (b).
E)Both (a)and (c).

Question 67

Which of the following is not part of the classical critique of quantitative approaches?

A)The objection that there is more to psychological phenomena than can be conveyed by mere numbers.
B)The objection that there is more to psychological phenomena than can be conveyed by crude attempts to manipulate discrete aspects of the environment one at a time.
C)The objection that most statistical analyses violate statistical assumptions.
D)Both (a)and (b)
E)All of the above (i.e., all are part of the classical critique).

Question 68

"Statistics that involve procedures that are relatively free of restrictive assumptions about the shape of the distribution of the population that a sample is drawn from." What construct is this a glossary definition of?

A)Distribution-free statistics.
B)Free-range statistics
C)Assumption free statistics.
D)Population-free statistics.
E)None of the above.

Question 69

Morten Fagerland has observed that in some areas of medical research the use of non-parametric statistical tests has increased in recent years? To what does he attribute this?

A)The fact that with larger sample sizes non-parametric tests tend to produce smaller p-values.
B)The fact that with larger sample sizes non-parametric tests tend to produce larger p-values.
C)The fact that with smaller sample sizes non-parametric tests tend to produce smaller p-values.
D)The fact that with smaller sample sizes non-parametric tests tend to produce larger p-values.
E)The fact that medical researchers have become increasingl alzy over time.

Question 70

The chi-squared test of independence involves analysing what form of data?

A)Data in which one variable has two or more levels.
B)Data in which two variables have two or more levels.
C)Data in which one variable has three or more levels.
D)Data in which two variables have three or more levels.
E)None of the above.

Question 71

"A distribution-free test that is used to compare the central tendency of sets of data that are collected in pairs from the same participants.In this regard it is the distribution-free analogue of a within-subjects t-test." What construct is this a glossary definition of?

A)Kruskal-Wallis test
B)Wilcoxon matched-pairs signed-ranks test
C)Chi-square test
D)Mann-Whitney test
E)McNemar test.

Question 72

"An assessment of the correspondence between an observed pattern of data and a particular model of that data." What statistical construct is this a glossary definition of?

A)Analysis of Variance (AVOVA).
B)Correspondence inference.
C)Goodness of fit.
D)Chi-square.
E)Base rate.

Question 73

A sports psychologist is seeking a screening task for talent identification that does not favour one gender over the other.He recruits a sample of 119 athletes (70 male and 49 female)and records the time taken to complete the task.The distribution has a very long tail so he decides to conduct a nonparametric test in SPSS.The output is shown below: Which of the following statements is false?

A)The Mann-Whitney test has no assumptions.
B)The Mann-Whitney test has less restrictive assumptions than the t-test.
C)A t-test could have been calculated using the same data
D)The decision to use a Mann-Whitney test can be justified in terms of the assumption of normality that applies to the t-test.
E)The sum of the mean ranks is approximately 119.

Question 74

"Statistical procedures that simultaneously examine relationships between two or more variables.Bivariate statistics - that examine the relationships between two variables - constitute a particular class of these".What type of statistics is this a glossary definition of?

A)Distribution-free statistics.
B)Non-parametric statistics.
C)Multivariate statistics.
D)Multi-correlational statistics.
E)None of the above.

Question 75

Which of the following statements is true?

A)Psychologists rarely encounter the problem of missing data.
B)When they encounter missing data, most psychologists are confused about how to handle it.
C)It is best to ignore the problem of missing data.
D)It is best not to discuss the problem of missing data.
E)None of the above.

Question 76

A sports psychologist is seeking a screening task for talent identification that does not favour one gender over the other.He recruits a sample of 119 athletes (70 male and 49 female)and records the time taken to complete the task.The distribution has a very long tail so he decides to conduct a nonparametric test in SPSS.The output is shown below: Which of the following statements is true?

A)The long tail in the distribution suggests that because some people had difficulty doing the task this means the task is not fair.
B)If some people had difficulty doing the task the research is not ethical.
C)Task difficulty does not itself make the research unethical it is a matter of whether the research violates ethical principles.
D)The long tail in the distribution suggests that because some people had difficulty doing the task this means the task is not fair but the research itself is ethical.
E)The research is ethical because sports psychology is a science and science is part of society.

Question 77

"A distribution-free test that is used to compare the central tendency of sets of data that are collected in pairs from the same participants.It is particularly applicable to 'before' and 'after' designs in which data are obtained from the same source on two occasions." What construct is this a glossary definition of?

A)Kruskal-Wallis test
B)Wilcoxon matched-pairs signed-ranks test
C)Chi-square test
D)Mann-Whitney test
E)McNemar test.

Question 78

A researcher has a data set in which there are two variables with two or more levels.Which of the following might be addressed using a distribution-free test?

A)Whether the two variables are related.
B)Whether the central tendency of the two variables is different.
C)Whether the central tendency of the two variables is the same.
D)Both (a)and (b).
E)All of the above.

Question 79

A clinical psychologist has conducted a test of a new therapeutic treatment for anxiety disorders in which the new therapy is compared against a no-treament control condition.Colleagues point out that more of the participants are dropping out of the control condition to seek treatment elsewhere.To see whether this is the case, the researcher collects a new sample of 200 people experiencing anxiety disorders.She then randomly assigns them to either a no-treatment control (Cond.1)or a treatment (Cond.2)condition and records whether participants stayed with the program (Gp 1)or left early (Gp 2).The results of a chi-square test are shown below (with output taken from Preacher's, 2001, on-line tool). Which of the following statements is true?

A)The program has not worked as p = .122.
B)The program has not worked because p > .15
C)The result does not tell us whether the program worked.
D)The program has worked because chi-square > 2.
E)None of the above.

Question 80

Which of the following statements is true?

A)There are no useful qualitative research methods in psychology.
B)There are no good qualitative research methods in psychology.
C)There are only six recognized qualitative research methods in psychology.
D)There are over 40 recognized qualitative research methods in psychology.
E)The main value of qualititative research is that it is cheap.