Deck 15: Statistical Evaluation of Data

Question 1

Which of the following is the definition of variance?

A)the sum of the scores divided by the number of scores
B)the average distance from the mean
C)the square root of the average distance from the mean
D)the average squared distance from the mean

Answer 1

D

Question 2

Statistical methods that use sample data to answer general questions about a population are called

A)parameters.
B)statistics.
C)descriptive statistics.
D)inferential statistics.

Answer 2

D

Question 3

A researcher is interested in the sleeping habits of students at the local state college.The average number of hours spent sleeping each night for the entire set of students enrolled at the college is an example of a

A)statistic.
B)parameter.
C)sample.
D)population.

Answer 3

B

Question 4

The most commonly used descriptive statistics for a set of numerical scores are

A)the mean and the variance.
B)the mean and the standard deviation.
C)the mean and the mode.
D)the mean and the median.

Question 5

Last year the college basketball team averaged 72 points per game with a standard deviation of 6.In the first game this season,they scored 85 points.Compared to last year,this score is

A)a little above average.
B)far above average.
C)above average,but it is impossible to describe how much above average.
D)There is not enough information to compare this score with last year's average.

Question 6

What is measured and described by the standard deviation?

A)the sum of the scores divided by the number of scores
B)the average distance from the mean
C)the square root of the average distance from the mean
D)the average squared distance from the mean

Question 7

A researcher who is interested in studying how humidity affects the eating behavior of rats obtains 25 rats to participate in a research study.The group of 25 rats is called a

A)statistic.
B)parameter.
C)sample.
D)population.

Question 8

A researcher has measured self-esteem for a sample of 40 first-grade children.If the researcher wanted to divide the children into two equal-sized groups consisting of the high self-esteem kids and the low self-esteem kids,which statistical measure would identify the correct score to serve as the boundary between the high and the low self-esteem groups?

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

Question 9

A researcher interested in language development obtains a sample of 25 three-year-old girls and a sample of 25 three-year-old boys.Each child is given a vocabulary test and the researcher computes the mean score for each sample.The difference between the two sample means is an example of a

A)parameter.
B)statistic.
C)variable.
D)constant.

Question 10

If a set of scores has a mean of 45 and a standard deviation of 9,then the scores have a variance of

A)3.
B)4.5.
C)18.
D)81.

Question 11

For scores measured on a nominal scale of measurement (for example,job classification),which measure of central tendency is appropriate?

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

Question 12

To construct a frequency distribution graph displaying a set of scores that were measured on an interval or a ratio scale,you should use

A)only a histogram.
B)only a polygon.
C)only a bar graph.
D)either a histogram or a polygon.

Question 13

Statistical techniques that summarize,organize,and simplify data are classified as

A)population statistics.
B)sample statistics.
C)descriptive statistics.
D)inferential statistics.

Question 14

A researcher is interested in the sleeping habits of students at the local state college.The entire set of students enrolled at the college is an example of a

A)statistic.
B)parameter.
C)sample.
D)population.

Question 15

The data for a correlation are displayed in a graph and show a set of points tightly clustered around a line sloping up to the right.Which of the following is the best estimate for the correlation?

A)+.10
B)+.90
C)- .10
D)- .90

Question 16

It is possible for a distribution of scores to have more than one

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

Question 17

Where is a score of X = 45 located in a distribution with a mean of 40 and a standard deviation of 12?

A)slightly above average
B)far above average
C)slightly below average
D)far below average

Question 18

To construct a frequency distribution graph displaying a set of scores that were measured on a nominal or an ordinal scale,you should use

A)only a histogram.
B)only a polygon.
C)only a bar graph.
D)either a histogram or a polygon.

Question 19

The concept of degrees of freedom is associated with

A)the mean for a sample.
B)the median for a sample.
C)the mode for a sample.
D)the variance for a sample.

Question 20

If a distribution of numerical scores contains one or two extremely high values,compared to the rest of the scores,which measure of central tendency would produce the most representative value?

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

Question 21

Without some correction,measures of inter-rater reliability

A)tend to overestimate the true level of agreement between the raters.
B)tend to underestimate the true level of agreement between the raters.
C)provide a very inaccurate estimate of the true level of agreement between the raters.
D)Inter-rater reliability tends to be accurate without correction.

Question 22

Values such as Cohen's d or r² (the percentage of variance accounted for)are used to measure

A)the validity of a hypothesis test.
B)the risk of committing a type I error.
C)the size of a treatment effect or the strength of a correlation.
D)the level of significance.

Question 23

Of the following research statistics,which one is not designed to correct problems with the calculation of split-half reliability?

A)Cohen's kappa
B)Cronbach's alpha
C)Kuder-Richardson formula 20
D)Spearman-Brown formula

Question 24

A researcher reports a significant correlation using an alpha level of .01.In this situation

A)the probability that the researcher is making a Type I error is p > .01.
B)the probability that the researcher is making a Type I error is p < .01.
C)the probability that the researcher is making a Type II error is p > .01.
D)the probability that the researcher is making a Type II error is p < .01.

Question 25

What outcome is likely to occur for a hypothesis test evaluating a treatment that has a very small effect?

A)a Type I error
B)a Type II error
C)correctly rejecting the null hypothesis
D)correctly failing to reject the null hypothesis

Question 26

If a sample is selected from a population,the sample mean is essentially guaranteed to be different from the population mean.In general,the naturally occurring difference between a sample statistic and the corresponding population parameter is called

A)type I error.
B)type II error.
C)sampling error.
D)inferential error.

Question 27

For a study comparing the proportion of registered Democrats who approve of the president's economic policy with the corresponding proportion for registered Republicans,the appropriate hypothesis test is

A)independent-measures t test.
B)repeated-measures t test.
C)single-factor analysis of variance.
D)chi-square test for independence.

Question 28

In a hypothesis test,standard error measures

A)the size of the treatment effect.
B)variability for the sample data.
C)the likelihood of a Type I or Type II error.
D)the amount of difference between the sample and the population that is reasonable to expect just by chance.

Question 29

By convention,it has been determined that alpha levels should be set no larger than

A).50.
B).10.
C).05.
D).01 .

Question 30

Because measures of split-half reliability are based on only half the items in the test,they tend to underestimate the true reliability of the whole test.Which of the following is an attempt to correct this problem?

A)the Spearman-Brown formula
B)the Kuder-Richardson formula 20
C)Cronbach's alpha
D)All of the other options are techniques to correct the underestimation problem.

Question 31

A Type II error means that a researcher has

A)falsely concluded that a treatment has an effect.
B)correctly concluded that a treatment has no effect.
C)falsely concluded that a treatment has no effect.
D)correctly concluded that a treatment has an effect.

Question 32

If the data in a scatter plot form a nearly perfect circle,the Pearson correlation would be approximately

A)+0.8 or +0.9.
B)+0.8 or +0.9.
C)1.00.
D)zero.

Question 33

For a within-subjects study evaluating the mean difference between two treatment conditions,the appropriate hypothesis test is

A)independent-measures t test.
B)repeated-measures t test.
C)single-factor analysis of variance.
D)chi-square test for independence.

Question 34

For a research study comparing the average problem-solving ability for 5-year-old,6-year old,and 7-year-old boys,the appropriate hypothesis test is

A)independent-measures t test.
B)repeated-measures t test.
C)single-factor analysis of variance.
D)chi-square test for independence.

Question 35

Without some correction,split-half reliability

A)tends to overestimate the true reliability of the full test.
B)tends to underestimate the true reliability of the full test.
C)tends to be an inaccurate estimate of the true reliability of the full test.
D)split-half reliability tends to be accurate without correction.

Question 36

The purpose of a hypothesis test is

A)to measures the strength of the relationship between two variables.
B)to determine the direction of the relationship between two variables.
C)to determine whether the patterns observed in sample data justify a conclusion that there are corresponding patterns in the population.
D)all of the other options are goals of a hypothesis test.

Question 37

A Type I error means that a researcher has

A)falsely concluded that a treatment has an effect.
B)correctly concluded that a treatment has no effect.
C)falsely concluded that a treatment has no effect.
D)correctly concluded that a treatment has an effect.

Question 38

What outcome is likely to occur for a hypothesis test evaluating a treatment that has a very large and robust effect?

A)a Type I error
B)a Type II error
C)correctly rejecting the null hypothesis
D)correctly failing to reject the null hypothesis

Question 39

One reason for reporting measures of effect size is that a very small treatment effect observed in a sample can be statistically significant,especially if

A)the sample size is very large.
B)the sample size is very small.
C)the sample is not representative of the population.
D)the sample is representative of the population.

Question 40

Of the following research statistics,which one is designed to correct for chance agreement between two raters when measuring inter-rater reliability?

A)Cohen's kappa
B)Cronbach's alpha
C)Kuder-Richardson formula 20
D)Spearman-Brown formula

Question 41

Without some correction,the percentage of agreement between two observers will tend to overestimate the true level of inter-rater reliability.

Question 42

In general,a sample with large variance is more likely to produce a significant result than a sample with small variance.

Question 43

One concern with measuring split-half reliability is that many different correlations are possible depending on how you decide to split the test items in half.

Question 44

The goal of a hypothesis test is to demonstrate that the patterns observed in the sample data represent real patterns in the population and are not simply due to chance or sampling error.

Question 45

The goal of descriptive statistics is to simplify,summarize,and organize data.

Question 46

The mode is the best way to measure central tendency for data from a nominal scale of measurement.

Question 47

A Type I error means that the hypothesis test has failed to detect a real treatment effect.

Question 48

A distribution of scores and a mean of 55 and a standard deviation of 4.The variance for this distribution is 16.

Question 49

It is always possible that patterns observed in sample data do not accurately represent patterns in the population.

Question 50

The median is the most commonly used measure of central tendency.

Question 51

In a distribution with a mean of M = 76 and a standard deviation of SD = 7,a score of 91 would be considered an extreme value.

Question 52

Cohen's kappa is used to correct for chance agreements when computing inter-rater reliability.

Question 53

A summary value,usually numerical,that describes a sample is called a parameter.

Question 54

A researcher records the average age for a group of 25 preschool children selected to participate in a research study.The average age is an example of a statistic.

Question 55

Without some correction,such as the Spearman-Brown formula,a correlation measuring split-half reliability will tend to overestimate the reliability for the full test.

Question 56

A negative correlation means that as the X values decrease,the Y values also tend to decrease.

Question 57

In a distribution with a mean of M = 36 and a standard deviation of SD = 8,a score of 40 would be considered an extreme value.

Question 58

A researcher reports a significant correlation using an alpha level of .01.The probability that the researcher is making a Type I error is p < .01.

Question 59

When a research report demonstrates a significant treatment effect at the .05 level,you can be more confident that the effect is real than if the researcher had reported a significant effect with an alpha level of only .01.

Question 60

In general,a correlation obtained from a large sample is more likely to be significant than a correlation obtained from a small sample.

Question 61

Explain the concept of statistical significance.Be sure that the concept of an alpha level in included in your explanation.

Question 62

Identify the two basic concerns with using a correlation to measure split-half reliability and explain how these concerns are addressed by Spearman-Brown,K-R 20,and Cronbach's alpha.

Question 63

Explain why a measure of effect size is recommended to accompany the result from a hypothesis test.

Question 64

Describe the general purpose for descriptive statistical techniques and for inferential statistical techniques.

Question 65

Identify the basic concern with using the percentage of agreement as a measure of inter-rater reliability and explain how this concern is addressed by Cohen's kappa.

Question 66

Identify the purpose for a hypothesis test and briefly explain how a hypothesis test accomplishes its purpose.

Question 67

Describe what is measured by the mean and what is measured by the standard deviation for a set of scores.Describe the set of scores in a sample with a mean of M = 40 and a standard deviation of SD = 4.(Where are most of the scores located? Where are they centered? What range of values do they cover?)