Deck 17: Statistical Reasoning: An Introduction

Question 1

The null hypothesis suggests that the two samples come from ___________ distribution(s), and the experimental hypothesis suggests that the two samples come from _____________ distribution(s).

A) different; different
B) different; the same
C) the same; different
D) the same; the same

Answer 1

the same; different

Question 2

In comparison with a confidence level of 0.05, a confidence level of 0.01 means

A) you are less sure that the null hypothesis is wrong.
B) you are more sure that the null hypothesis is wrong.
C) you are absolutely sure that the null hypothesis is wrong.
D) you are more sure that the experimental hypothesis is wrong.

Answer 2

you are more sure that the null hypothesis is wrong.

Question 3

The mean is

A) the arithmetic average.
B) the middle score.
C) the average of the highest and lowest scores.
D) the most common score.

Answer 3

the arithmetic average.

Question 4

If the mean I.Q. score is 100 and the standard deviation of I.Q. scores is 15, then an I.Q. of 130 will have a z-score (or standard score) of

A) 1.00.
B) 0.00.
C) 2.00.
D) -2.00.

Question 5

Inferential statistics allow you to decide whether a difference between the experimental and the control group is due to _______________ or ________________.

A) manipulation; chance
B) manipulation; experimental error
C) sampling error; independent variable
D) independent variable; experimental error

Question 6

The ___________ is the standard deviation of sample means.

A) absolute mean deviation
B) central dendency
C) point biserial
D) standard error

Question 7

A conservative statistical test is one that

A) minimizes both Type I and Type II errors.
B) minimizes Type I errors, but increases the chance of Type II errors.
C) minimizes Type II errors, but increases the chance of Type I errors.
D) increases the chance of both Type I and II errors.

Question 8

Failing to reject the null hypothesis when it is actually true is a

A) Type I error.
B) Type II error.
C) reasonable thing to do.
D) result of an inappropriate level of confidence.

Question 9

The two properties about a set of measurements of a dependent variable that we are most interested in describing are

A) frequency and average.
B) average and correlation.
C) central tendency and dispersion.
D) histograms and polygons.

Question 10

An advantage of the median over the mean is that the median is

A) easier to calculate.
B) more sensitive to each of the sample scores.
C) a measure of dispersion.
D) less influenced by extreme scores..

Question 11

The ________________ is the sum of all the scores divided by the number of scores.

A) median
B) mean
C) mode
D) standard deviation

Question 12

A score that lies one standard deviation above the mean

A) has the same X value as the point of inflection.
B) has a z-value of +1.0.
C) has a z-value of -1.0.
D) has a z-score equal to the mean.

Question 13

All of the following statements are true of the normal distribution EXCEPT

A) the mean and median coincide.
B) the most common scores lie near the mean.
C) the more rare the score is, the further from the mean it lies.
D) nearly all the scores lie outside of two standard deviations of the mean.

Question 14

The ______________ divides a distribution in half.

A) range
B) mean
C) mode
D) median

Question 15

The standard deviation is

A) the square of the variance.
B) the square root of the variance.
C) smaller than the mean.
D) the difference between the highest and lowest scores.

Question 16

The generally preferred measure of central tendency is usually the

A) range.
B) mean.
C) standard deviation.
D) median.

Question 17

Failing to reject the null hypothesis when it is actually false is a

A) Type I error.
B) Type II error.
C) standard error.
D) silly thing to do.

Question 18

Which of the following is the most useful descriptive statistic for measuring dispersion?

A) range
B) variance
C) mean deviation
D) standard deviation

Question 19

Rejecting the null hypothesis when it is actually false is

A) a Type I error.
B) a Type II error.
C) more likely with two-tailed tests.
D) a correct decision.

Question 20

As the sample size (n) increases,

A) the standard error of the mean increases.
B) the power of the statistical tests decreases.
C) the probability of the Type II error increases.
D) the power of the statistical tests increases.

Question 21

Inferential statistics are used to assess the importance of experimental results.

Question 22

The Pearson product?moment correlation coefficient is referred to as

A) z-score.
B) s.d.
C) r.
D) p.c.c.

Question 23

In comparison with one-tailed tests, two-tailed tests are

A) less conservative.
B) more conservative.
C) less likely to lead to a Type II error.
D) more likely to lead to a Type I error.

Question 24

Simple analysis of variance is used in designs having

A) one independent variable.
B) more than one independent variable.
C) more than one independent variable (IV) but less than four IVs.
D) more than one dependent variable.

Question 25

The Mann-Whitney U test is

A) nonparametric.
B) distribution free.
C) designed for two groups of scores.
D) all of these

Question 26

In describing an array of data, psychologists typically present two descriptive statistics, the mean and the variance.

Question 27

If a person's score is part of a normal distribution of scores, we can determine the relative rank of that score if we know

A) the person's score.
B) the mean of the distribution.
C) the standard deviation of the distribution.
D) we must know all of these

Question 28

The larger the sample, the more we can trust the sample ___________ to be an estimate of the population ___________.

A) mean; mean
B) standard error of the mean; standard deviation
C) standard error of the mean; variance
D) difference; mean

Question 29

The Wilcoxon signed-ranks test

A) is used for between-subjects designs.
B) takes into account both the magnitude and the direction of the difference between paired scores.
C) cannot be used for within-subjects designs.
D) is a parametric test.

Question 30

Which of the following is a parametric test?

A) Mann-Whitney U Test
B) sign test
C) t test
D) Test of Independence

Question 31

The power of a statistical test refers to its ability to

A) reject false null hypotheses.
B) reject false experimental hypotheses.
C) reject true null hypotheses.
D) reject true experimental hypotheses.

Question 32

The median is often used because it is insensitive to extreme scores.

Question 33

Which of the following is NOT a measure of dispersion?

A) range
B) standard deviation
C) variance
D) median

Question 34

In a multifactor analysis of variance, how the variables affect one another is called

A) a main effect.
B) an interaction effect.
C) an experimental effect.
D) a variance effect.

Question 35

The inflection point in the normal curve is _____________ from the mean.

A) less one standard deviation
B) two standard deviations
C) three standard deviations
D) one standard deviation

Question 36

In a histogram, the values for the dependent variable are displayed on the ordinate (Y-axis) and the frequency with which each value of the dependent value occurred is displayed on the abscissa (X-axis).

Question 37

In hypothesis testing, both the magnitude of the difference and the direction of the difference between two groups are considered in

A) two-tailed tests.
B) one-tailed tests.
C) parametric tests.
D) nonparametric tests.

Question 38

The F statistic is

A) the ratio of between-group variance to within-group variance.
B) between-groups variance minus within-groups variance.
C) within-groups variance minus between-groups variance.
D) a measure of error variance.

Question 39

The number of participants in a study is denoted by

A) s.
B) n.
C) z.
D) r.

Question 40

If experimental treatments produce an effect,

A) between-groups variance will increase.
B) between-groups variance will decrease.
C) within-groups variance will increase.
D) within-groups variance will decrease.

Question 41

Approximately 68 percent of the scores in a normal distribution have z values between -2.0 and +2.0.

Question 42

If the null hypothesis is true, alpha equals the type I error probability.

Question 43

The variance of a distribution is defined as the sum of the squared deviations from the mean divided by the number of scores.

Question 44

Characteristics of a population of scores are called parameters, while characteristics of a sample of scores drawn from a larger population are called statistics.

Question 45

The absolute mean deviation is an adequate measure of dispersion, and it is preferable over the standard deviation and variance because of its usefulness in inferential statistics.

Question 46

Most hypothesis testing is performed using known population parameters.

Question 47

As the level of significance is changed from .01 to .05, the probability of making a type I error increases and the probability of making a type II error decreases.

Question 48

Two-tailed tests are more liberal and less powerful than one?tailed tests.

Question 49

The degrees of freedom equals the number of scores that are free to vary minus one.

Question 50

The standard error of the mean is the standard deviation of a distribution of sample means.

Question 51

The standard error of the mean is the square root of the mean.

Question 52

The t test is a nonparametric test.

Question 53

It is impossible to prove the alternative hypothesis to be true.

Question 54

The Mann-Whitney U test can only be used for between-subject designs and the Wilcoxon signed-ranks test can only be used for related measures designs.

Question 55

If an experiment involving an experimental group and a control group were performed over and over and the distribution of the difference between the means of the experimental and control groups was plotted, this distribution would approximate a normal distribution.

Question 56

The standard error of the mean increases with increasing sample size.

Question 57

After conducting an F test, eta is used to determine how wrong the null hypothesis is.

Question 58

If the null hypothesis is true, F should equal 1.0.

Question 59

The power of a test is the probability of rejecting the null hypothesis when it is actually true.

Question 60

Approximately 96 percent of the scores in a normal distribution have z values between -2.0 and +2.0.

Question 61

Describe three ways that statistics can be misused to create a false impression.

Question 62

What is the difference between inferential statistics and descriptive statistics? Give two examples of each type.

Question 63

What are the properties of the normal curve? What does knowing these properties do for us? What is a standard score?

Question 64

Why should one try to maximize the sample size?

Question 65

What are standard scores? How are they calculated? Why are they useful?

Question 66

Discuss the various measures of dispersion (i.e., absolute mean deviation, variance, and standard deviation). Which of them do you think is the most useful? Why?

Question 67

What is the difference between parametric and nonparametric tests? Give two examples of each type.

Question 68

What are type I and type II errors? Under what conditions would you be willing to risk a type I error?

Question 69

What is power? What are the factors that influence the power of a statistical test?

Question 70

Manipulation of type of instructions given is an example of quantitative variation of an independent variable.

Question 71

If the null hypothesis is true, F should be 1.0. If the null hypothesis is false, F should be greater than 1.0. However, sometimes Fs less than 1.0 are observed. Explain the logic behind the F test and how Fs less than 1.0 can be obtained.