SOCL 2201 FINAL

Which of the following is not a requirement of a two-way chi-square? a. A comparison between two or more mean scores b. Nominal data c. Random sampling d. The expected cell frequencies should not be too small

a) a comparison between two or more mean scores

When respondents who score high on the SATs tend also to get high grades in college, whereas those who score low on the SATs do poorly in college, there is reason to posit a. a positive correlation between SATs and college grades. b. a negative correlation between SATs and college grades. c. a zero correlation between SATs and college grades. d. None of the above

a) a positive correlation between SATs and college grades

A partial correlation coefficient gives us the correlation between two variables a. after removing the common effect of a third variable. b. at the ordinal level of measurement. c. after introducing the common effect of a third variable. d. when we are partially certain that no relationship between them actually exists.

a) after removing the common effect of a third variable

In a chi-square test, the expected frequencies: a. are expected to occur under the terms of the null hypothesis. b. are expected to occur under the terms of the research hypothesis. c. refer to those frequencies actually observed from the results of conducting research. d. are never known by the researcher

a) are expected to occur under the terms of the null hypothesis

At P = .01, a mean difference in a sampling distribution occurs: a. by sampling error once in every 100 mean differences. b. by sampling error more than 99 times in every 100 mean differences. c. very frequently by sampling error.

a) by sampling error once in every 100 means

Rather than a series of t tests, analysis of variance is used because: a. it holds Type I error at a constant level. b. it is too much work to do a series of t tests. c. it increases Type I error. d. it makes a number of decisions, whereas a series of t tests makes a single overall decision.

a) it holds Type I error at a constant level

The level of probability at which the null hypothesis can be rejected with confidence is known as the: a. level of significance. b. distribution. c. degrees of freedom. d. All of the above

a) level of significance

When you predict that a result will be in one or another direction (for example, that a particular treatment will reduce recidivism), you would use a: a. one-tailed test. b. two-tailed test. c. three-tailed test. d. no-tailed test

a) one-tailed test

Interpret the following correlation statistics with respect to strength and direction: a. r = 0.951 b. r = -0.232 c. r = 0.096 d. r = -0.738 e. r = 0.415

a) strong positive b) weak negative c) weak positive d)strong negative e)moderate positive

The exact probability of obtaining an outcome when the null hypothesis is true is known as: a. the a level. b. Type I error. c. the P value. d. Type II error

a) the a level

The size of the tail regions under the curve that will lead us to reject the null hypothesis is (are): a. the alpha value. b. the P value. c. the standard error of the difference. d. the degrees of freedom.

a) the alpha value

To find a significant difference with an analysis of variance, you hope to maximize: a. the between-groups mean square. b. the within-groups mean square. c. the within-groups sum of square. d. variation within groups.

a) the between-groups mean square

Which of the following is not required by Pearson's r? a. A straight-line relationship b. Nominal data c. Random sampling d. Normally distributed characteristics 6

b) Nominal data

In regression analysis, we predict the value of Y from X based on a a. significant correlation coefficient. b. a mathematical equation. c. a weak positive correlation. d. a normal curve.

b) a mathematical equation

The purpose of a partial correlation is to: a. test for a significant relationship b. control a relationship for a third variable c. make curvilinear relationship into a linear one d. calculate a correlation for part of a sample

b) control a relationship for a third variable

The larger the z score: a. the greater the distance of our mean difference from zero. b. the more likely we are to reject the null hypothesis. c. the more likely that our result is significant. d. All of the above

d) all of the above

Which of the following is not a requirement for employing a t test? a. A comparison between two means b. Random sampling c. A normal distribution d. Nominal data

d) nominal data

Which of the following is not a requirement of analysis of variance? a. A comparison of three or more independent means b. Random sampling c. A normal distribution d. Ordinal data

d) ordinal data

The chi-square test compares: a. the difference in means for two groups at a time. b. the difference in means for more than two groups. c. the variance of one group with the variance of others. d. the difference between expected and observed frequency counts.

d) the difference between expected and observed frequency counts

A two-tailed significance test suggests that: a. the resulting outcome has a particular direction (that is, a clinical treatment produces a decrease in symptoms). b. the resulting outcome is unlikely. c. the resulting outcome is likely. d. the resulting outcome may be in either direction (that is, a clinical treatment produces either an increase or decrease in symptoms).

d) the resulting outcome may be in either direction (that is, a clinical treatment produces either an increase or decrease in symptoms).

Which of the following denotes the slope or the regression coefficient for X? a. The Y-intercept b. The term a c. The error term d. The term b

d) the term b

The analysis of variance produces an F ratio in which the denominator represents an estimate of: a. the total sum of squares. b. the total variation. c. variation between groups. d. variation within groups.

d) variation within groups

To employ the median test, you must be able to: a. compare the means. b. assume a normal distribution. c. compare three or more independent samples. d. rank-order a set of cases.

d)rank-order a set of class

If a calculated F ratio is larger than the value in Table D, it means there is: a. a significant difference between groups and the null hypothesis can be rejected. b. a significant difference between groups and the null hypothesis can be retained. c. not a significant difference between groups and the null hypothesis can be rejected. d. not a significant difference between groups and the null hypothesis can be retained.

a

The median test should be used to: a. determine the likelihood that a sample's median is actually different from "0." b. determine the likelihood that two or more random samples have been taken from populations that have the same median. c. determine the likelihood that the observed frequency counts in a nominal-level variable for one group are different from another nominal-level variable. d. investigate the probability that the sample standard deviation does not accurately reflect a population standard deviation.

b) determine the likelihood that two or more random samples have been taken from populations that have the same median.

For an F ratio to be statistically significant, the between-groups mean square must be: a. greater than the total sum of squares. b. greater than the within-groups mean square. c. less than the within-groups mean square. d. less than the within-groups sum of squares.

b) greater than the within-groups mean square

Using two or more predictors, we conduct a a. simple regression. b. multiple regression. c. partial correlation. d. Pearson's correlation.

b) multiple regression

In a chi-square analysis, the larger the difference between expected and observed frequencies, the more likely you are to a. retain the null hypothesis. b. reject the null hypothesis. c. instead use a t ratio or some other parametric test.

b) reject the null hypothesis

The larger the calculated t value, the more likely we are to a. retain the null hypothesis. b. reject the null hypothesis. c. conclude that our result is not statistically significant.

b) reject the null hypothesis

When the points in a scatter plot cluster closely around a straight line, the correlation can be said to be: a. weak. b. strong. c. positive. d. negative.

b) strong

The F ratio is larger when: a. the between-groups mean square is smaller. b. the within-groups mean square is smaller. c. the difference between means is smaller. d. None of the above

b) the within-groups mean square is smaller

Retaining the null hypothesis assumes which of the following? a. There is no difference between sample means. b. There is no difference between population means. c. The difference between means is significant. d. The difference between means is too large to be a sampling error

b) there is no difference between population means

A measure of the proportion of the variance in Y explained by X is known as a. the coefficient of nondetermination. b. the coefficient of determination. c. multiple regression. d. the slope.

b) to coefficient to determination

The farther out in the tail of a distribution our critical value falls, the greater the risk of making a a. Type I error. b. Type II error. c. Type I and Type II errors. d. Type III error.

b) type II error

A multiple comparison of means is necessary when the analysis of variance results in: a. a significant difference between two means. b. a nonsignificant difference between two means. c. a significant difference among three or more means. d. a nonsignificant difference among three or more means.

c) a significant difference among three or more means.

A correlation coefficient expresses in a single number a. the strength of a correlation. b. the direction of a correlation. c. both the strength and the direction of a correlation. d. None of the above

c) both the strength and the direction of a correlation

The standard error of the difference between means is defined as: a. the standard deviations of the two samples we have drawn. b. the standard deviations in the two populations from which two samples were drawn. c. our estimate of the standard deviation of the theoretical sampling distribution of differences between means.

c) our estimate of the standard deviation of the theoretical sampling distribution of differences between means

In a statistical sense, "significant" means a. important. b. large. c. real. d. unlikely to be found in the population.

c) real

The two-way chi-square might be used as a nonparametric alternative instead of ________ when comparing two groups. a. confidence intervals b. standard deviation c. t ratio d. analysis of variance

c) t ratio

The increase or decrease in Y expected with each unit change in X is known as a. the Y-intercept, a. b. the error term, e. c. the regression coefficient, b. d. PRE.

c) the regression coefficient, b.

Which is not true of parametric tests? a. They require interval data. b. They require normality in the population. c. They are less powerful than nonparametric tests. d. None of the above is true of parametric tests.

c) they are less powerful than nonparametric tests

r = -.17 represents a correlation that is: a. strong and negative. b. strong and positive. c. weak and negative. d. weak and positive.

c) weak and negative

For a chi-square value to be statistically significant, the following must be true: a. The median of one group must be nearly the same as the other groups. b. The mean of one group must be nearly the same as the other groups. c. The observed frequencies must differ from the expected frequencies by a sufficient amount. d. The observed frequencies must be nearly the same as the expected frequencies.

c)The observed frequencies must differ from the expected frequencies by a sufficient amount.