9) A test for equality of two variances is based on A. the ratio of the sample variances B. the difference between the population variances. C. C. the difference between the sample coefficients of variation. D. A. the difference between the sample variances.
A. the ratio of the sample variances
10) Which of the following statements is false for an F-distribution? A. Degrees of freedom for the numerator may be larger, smaller, or equal to the degrees of freedom for the denominator. B. The exact shape of the distribution is determined by two numbers of degrees of freedom. C. Degrees of freedom for the denominator are always smaller than degrees of freedom for the numerator. D. Variables that are F distributed range from 0 to infinity.
C. Degrees of freedom for the denominator are always smaller than degrees of freedom for the numerator.
12) In a test for the equality of two variances (two-tailed), when the populations are normal, a 5% level of significance was used. Sample sizes were n1 = 13 and n2 = 10. The upper critical value for the test is A. =FINV(1-0.025, 13, 10). B. =FINV(0.025, 12, 9). C. =FINV(0.025, 13, 10). D. =FINV(0.05, 12, 9).
B. =FINV(0.025, 12, 9).
13) The expected value of the sampling distribution of the sample mean equals the populations mean A. For all populations. B. When the population is symmetric. C. When the population size N>30. D. When the population is normally distributed.
A. For all populations
14) The standard error of the sample mean is equal to 5 when n=25. If the sample size increases by a factor of four, who will the standard error change? A. It will quadruple. B. It will be cut in half. C. It will be cut to ¼ of 5. D. It will double.
B. It will be cut in half.
15) For a sample size of 1, the sampling distribution of the mean will be normally distributed A. Only if the population is normally distributed. B. Only if the population values are larger than 30. C. Only if the shape of the population is positively skewed. D. Regardless of the shape of the population.
A. Only if the population is normally distributed.
16) A test for equality of two variances has sample sizes n1 = 13 and n2 = 10. The degrees of freedom for the test are A. 12 and 9. B. 13 and 10. C. 26. D. 21.
A. 12 and 9
17) You are conducting a two-tailed test of means, but your software package only calculates a one-tailed p-value equal to 0.13. The actual p-value for your test is A. You need a table to calculate this value. B. 0.26. C. 0.065. D. 0.13.
B. 0.26
18) When testing for differences between two means, the Behrens-Fisher problem arises when the sample populations are A. are non-normal and have unequal variances. B. are non-normal and have equal variances. C. are normal with unequal variances. D. normal with equal variances.
C. are normal with unequal variances
19) The purpose of designing a randomized block experiment is to reduce the between-treatments variation (SST) to more easily detect differences between the treatment means. A. True B. False
B. False
20) Main effects are simpler to interpret when the test of the interaction term is not significant in a two-factor ANOVA. A. True B. False
A. True
21) The F-test of the randomized block design of the analysis of variance has the same requirements as the independent sample design; that is, the random variable must be normally distributed and the population variances must be equal. A. True B. False
A. True
22) Which of the following statistics from the ANOVA table do not have an additive relationship? A. It is not possible to tell. B. Mean squares C. Degrees of freedom D. Sum of squares
B. Mean squares
23) Hartley's test measures the equality of the means for several groups. A. True B. False
B. False
24) Comparison of c means in a one-factor ANOVA is the same as using c repeated t-tests. A. True B. False
B. False
25) Statistical methods that require few assumptions, if any, about the population distribution are known as A. All of these B. free agent techniques. C. parametric techniques.D. nonparametric techniques.
. D. nonparametric techniques.
26) The nonparametric counterpart of the parametric one-way analysis of variance F-test is the A. Wilcoxon signed rank sum test. B. Spearman’s rho. C. Friedman test. D. Kruskal-Wallis test.
. D. Kruskal-Wallis test.
27) Consider the following data set: 14, 14, 15, 16, 18, 19, 19, 20, 21, 22, 23, 25, 25, 25, 25, and 28. The rank assigned to the four observations of value 25 is A. 13.5 B. 12 C. 12.5 D. 13
A. 13.5
28) Of the values for a chi-squared test statistic listed below, which one is most likely to lead to rejecting the null hypothesis in a goodness-of-fit test? A. 0 B. 45 C. 2.1 D. 1.2
B. 45
29) If each observation can be classified into one of several mutually exclusive and collectively exhaustive categories, the population is aA. binomial. B. normal C. multinomial.D. chi-squared.
C. multinomial
30) In the chi-squared goodness-of-fit test, if the expected frequencies ei and the observed frequencies fi were quite different, we would conclude that the A. chi-squared distribution is invalid, and we would use the t-distribution. instead B. null hypothesis is false, and we would reject it. C. null hypothesis is true, and we would not reject it. D. alternative hypothesis is false, and we would reject it.
B. null hypothesis is false, and we would reject it
31) The distribution for Spearman’s rho is A. z B. F C. t D. x2.
A. z
32) If each group has at least five observations, the distribution of the Kruskal-Wallis H is A. z B. F C. t D. x2
D. x2
33) The nonparametric counterpart of the randomized block model of the ANOVA is the A. Wilcoxon signed rank sum test. B. Kruskal-Wallis test. C. Friedman test. D. Wilcoxon rank sum test.
C. Friedman test
34) In regression analysis, if the coefficient of determination is 1.0, then the coefficient of correlation must be 1.0. A. True B. False
B. False
35) The value of the sum of squares for regression SSR can never be larger than the value of total sum of squares SST. A. True B. False
A. True
36) A regression analysis between sales, in $1,000, and advertising, in $100, resulted in the following least squares line: Sales' = 75 + 6*(Advertising). This implies that if advertising is $800, sales will be A. $487,500. B. $12,300. C. $4,875. D. $123,000.
D. $123,000.
37) The bivariate correlation coefficient always has the same sign as b1 in Y = b0 + b1X. A. True B. False
A. True
38) From the coefficient of multiple determination, we cannot detect the strength of the relationship between the dependent variable y and any individual independent variable. A. True B. False
A. True
39) What randomness exists in the linear regression model? A. The randomness from the explanatory variables, the X's B. The randomness from what is unexplained, the error C. The randomness of the dependent variable, the Y's D. None of these
D. None of these
40) Given that SSE = 84 and SSR = 358.12, the coefficient of correlation, also called the Pearson coefficient of correlation, must be 0.90. A. True B. False
B. False
41) Except for the values r = -1, 0, and 1, we cannot be specific in our interpretation of the coefficient of correlation r. However, when we square it, we produce a more meaningful statistic. A. True B. False
B. False
42) In a simple linear regression model, testing whether the slope ß1 of the population regression line could be zero is the same as testing whether or not the population coefficient of correlation p equals zero A. True B. False
A. True
43) If a group of independent variables are not significant individually but are significant as a group at a specified level of significance, this is most likely due to A. the presence of binary variables. B. multicollinearity. C. autocorrelation D. the absence of binary variables.
multicollinearity
44) If the Durbin-Watson statistic has a value close to 0, which assumption is violated? A. Independence of errors B. Normality of the errors C. Homoscedasticity D. None of these
Independence of errors
45) If the Durbin-Watson statistic, DW, has values greater than 2, this indicates A. a positive first–order autocorrelation. B. a negative first–order autocorrelation.C. no first–order autocorrelation at all. D. None of the above
B. a negative first–order autocorrelation
46) Smoothing time series data by the moving average method or exponential smoothing method is an attempt to remove the effect of the A. irregular variation component. B. cyclical component.C. trend component.D. seasonal component.
A. irregular variation component
47) The time series component that reflects variability over short, repetitive time periods that last less than one year is called A. irregular variation. B. cyclical variation.C. long–term trend.D. seasonal variation.
. D. seasonal variation.
48) If we want to measure the seasonal variations on stock market performance by quarter, we would need A. 1 indicator variables. B. 3 indicator variables. C. 4 indicator variables. D. 2 indicator variables.
B. 3 indicator variables
49) When no point lies outside the control limits of a chart, we conclude that variation in the process is A. due to common cause and there is not enough evidence to infer that the process is out of control. B. due to special causes and there is enough evidence to infer that the process is under control. C. due to special causes and there is not enough evidence to infer that the process is out of control. D. due to common cause and there is enough evidence to infer that the process is out of control.
A. due to common cause and there is not enough evidence to infer that the process is out of control.
50) In statistical process control, a Type I error occurs if we decide that the process is A. out of control when it is out of control. B. out of control when it is under control.C. under control when it is out of control.D. under control when it is under control.
. B. out of control when it is under control
51) Variations in process output that are caused by a number of randomly occurring events that are part of the production process are A. special causes. B. common causes. C. out-of-control causes. D. All of the these
B. common causes