Return to Brian Schott @ GSU

Return to the list of chapters

98-2 P (X < c) = .95?

105-2 the probability of obtaining an F-ratio exceeding

1278-1 The appropriate statistical test for this comparison is:

1430-1 chicken soup consumption between sexes statistically significant?

1536-1 Set 95% confidence limits for the difference between means

1612-1 F > 9.48773 and the computed value of F from the data is .86.

1614-2 and your P-value satisfies .01 < P < .05.

1650-1 What useful information can you supply future investigators when

1687-4 "In ANOVA, if we wish to investigate the difference among five"

1872-3 Suggest some factors that might influence variability in weights

1873-2 Data containing less variability than would be expected by chance

1895-1 You are to develop plans for a follow-up study.

1917-1 s**2 = (Sum of Squares)/(n-1)

1953-2 The degrees of freedom associated with a error (or within)

1953-3 the degrees of freedom associated with within sum of squares depend

2102-3 which of the following is an estimate of the variance of

2118-1 MEAN SQUARE

2548-1 "If known variation is large compared to unknown variation,"

2550-1 sees that his MS within groups is larger than his MS among groups.

2551-1 What decision would be made regarding H(0):

2552-1 Then we know that the experimenter used

2553-1 Items i. thru iii. are based on a teaching experiment involving

2567-1 what is the number of degrees of freedom for

2570-1 How many degrees of freedom do we have for the within groups sum of

2577-2 F = 3.5 so hypothesis is

2582-1 A fisheries researcher wishes to conclude that there is a difference in

2584-2 "If the sample means for each of the treatment groups were identical,"

2586-1 exceeding that reported in the F table

2588-1 -0.10 to 1.5. Which one of the following statements is true?

2591-1 Samples from 3 classes were given an identical math test.

2593-1 "If you reject, which shelving policies are different? "

2598-1 "the hypothesis that ""mothers are more likely to be success"

2601-1 """The wealthier a person, the more likely he will be relatively"

2603-1 hypothesis that male students are more assertive than female students

2609-1 Complete the ANOVA Table and test if all routes are equally

2612-1 What hypothesis is tested using this F ratio?

2702-1 Complete the following analysis of variance table:

2705-1 what is the value of (corrected) total sums of squares?

2706-1 what is the sum of squares for between groups?

2709-1 Consider each of the pairs of methods.

2719-4 indicates how a particular sum-of-squares value is calculated.

2723-4 "In a one-way ANOVA, the between mean square plus the within mean square"

2771-2 Are you satisfied with this conclusion? Why or why not?

2779-1 but use of an LSD is much less exposed to Type II errors.

2781-2 Which of the following pairs of sample means should NOT be judged

Return to the list of chapters

Return to Brian Schott @ GSU

Q: If X has an F distribution with df=4,5, what is the value of c where P (X < c) = .95?

Back to this chapter's Contents

Q: True or False? If False, correct it. In a one-way classification ANOVA, when the null hypothesis is false, the probability of obtaining an F-ratio exceeding that reported in the F table at the .05 level of significance is greater than .05.

Back to this chapter's Contents

Q: In a study, subjects are randomly assigned to one of three groups: control, experimental A, or experimental B. After treatment, achieve- ment test scores for the three groups are compared. The appropriate statistical test for this comparison is: a. the correlation coefficient b. chi square c. the t-test d. the analysis of variance

Back to this chapter's Contents

Q: A survey of 114 men and 126 women produced the result that the mean amount of chicken soup consumed by the men in a month's time was .67 liters, compared with a mean of .54 liters for the women. The var- iance of the chicken soup consumption for the men was 25% greater than that for the women. A 95% confidence interval for the difference between the means (men's mean - women's) was found to be -.07 to +.33 liters. At the .05 level, is the difference in mean chicken soup consumption between sexes statistically significant? a) no b) yes c) Can't tell from the data given

Back to this chapter's Contents

Q: The report for an investigation includes: Means for weight loss by dieters Cream puff diet 0 Frogurt diet 5 Beer and raw eggs diet 7 Saltines and kind words diet 8 HSD(.05) 2 Set 95% confidence limits for the difference between means for: a. Cream puff diet vs. beer and raw eggs b. Beer and raw eggs vs. saltines and kind words

Back to this chapter's Contents

Q: Suppose the critical region for a certain test of hypothesis is of the form F > 9.48773 and the computed value of F from the data is .86. (F refers to an F statistic.) Then: a) H(O) should be rejected. b) H(A) is two-tailed. c) The significance level is given by the area to the right of 9.48773 under the appropriate F distribution. d) None of these.

Back to this chapter's Contents

Q: Suppose H(O): MU(1) = MU(2) and your P-value satisfies .01 < P < .05. Which of the following conclusions can be drawn? (a) Fail to Reject H(O) because P is small. (b) Reject H(O) because P is small. (c) Fail to Reject H(O) because P is large. (d) Reject H(O) because P is large. (e) Fail to Reject H(O) at ALPHA = .01.

Back to this chapter's Contents

Q: A carefully designed experiment has just been concluded. Execution of the experiment was flawless. Unfortunately use of ALPHA (.05) indic- ated no significant differences among treatments. What useful information can you supply future investigators when you report on this experiement? Indicate agreement with yes or disagreement with no for each of the fol- lowing items. a. No useful information. These are negative results and there is nothing useful to report. b. The estimated variance (and its df) can be useful to future invest- igators. c. A careful description of experimental conditions and treatments may be useful as an indication of circumstances where responses are about the same. d. Significant differences can be reported by changing the Type I error rate from .05 to .10, .20 or whatever is needed to declare significance.

Back to this chapter's Contents

Q: True or False? If False, correct it. In ANOVA, if we wish to investigate the difference among five means, it is good statistical procedure to perform a t-test on each pair of means.

Back to this chapter's Contents

Q: Suggest some factors that might influence variability in weights measured on a group of men.

Back to this chapter's Contents

Q: True or False? If False, correct it. Data containing less variability than would be expected by chance are always preferable.

Back to this chapter's Contents

Q: A report on an investigation includes the following information related to the influence of several mouth washes to length of time that breath remains "great". Analysis of Variance Source df SS Mean Square Treatments 3 500 166.67 Error 196 19,600 100 Total 199 20,100 Means (hours that breath remained "great") Whiskey 12 Brand X 11 Water 9 Brand L 8 F(critical) (.05) = 3.92 with 196 df You are to develop plans for a follow-up study. In particular, you are to re-examine the difference between Brand X and Brand L. In looking at methods for estimating number of replicates needed you find that you need values for - 1. the size of the difference to be detected 2. the anticipated standard deviation 3. the anticipated variance On the basis of the above report what values will you use for each of the above 3 items? Why?

Back to this chapter's Contents

Q: Given the following observed number of pigs for 8 litters, the numerator of the formula for s**2 is called the corrected sum of squares as illus- trated. s**2 = (Sum of Squares)/(n-1) Find the sum of squares. X(1) = 9 X(5) = 10 X(2) = 6 X(6) = 7 X(3) = 14 X(7) = 8 X(4) = 9 X(8) = 9 a) 5 b) SQRT(5) c) 40 d) 80

Back to this chapter's Contents

Q: True or False? If False, correct it. The degrees of freedom associated with a error (or within) term in ANOVA depends only on sample size.

Back to this chapter's Contents

Q: True or false? IF FALSE, CORRECT IT. In ANOVA the degrees of freedom associated with within sum of squares depend on how many independent sample values were used and how many other independent parameters have been estimated.

Back to this chapter's Contents

Q: In a simple analysis of variance problem, which of the following is an estimate of the variance of individual measurements (after the various effects have been accounted for)? (MS means SS/df so each of answers is a Mean Square.) a. MS(between) b. MS(within) c. MS(total) d. none of the above

Back to this chapter's Contents

Q: Define the following term and give an example of its use. Your example should not be one given in class or in a handout. MEAN SQUARE

Back to this chapter's Contents

Q: The total variation in response, assuming no bias, is due to error (unexplained variation) plus differences due to treatments (known variation). If known variation is large compared to unknown variation, which of the following conclusions is the best? a) There is no difference in response due to treatments. b) There is a difference in response due to treatments. c) The treatments are not comparable. d) The cause of the response is due to something other than treatments.

Back to this chapter's Contents

Q: An investigator randomly assigns 30 college students into three equal size study groups (early morning, afternoon, late night) to determine if the period of the day at which people study has an effect on their retention. The students live in a controlled environment for one week, on the third day of which the experimental treatment (study of predetermined material) is administered. The seventh day the investigator tests for retention, and in computing his analysis he sees that his MS within groups is larger than his MS among groups. What is the indication of this result? a. An error in calculation was made. b. There was more than the expected variability between groups. c. There was more variability between subjects within the same group than there was between groups. d. That there should have been additional controls in the experi- ment.

Back to this chapter's Contents

Q: Source SS df ------ -- -- Between 30.5 4 Within Total 165.0 99 What decision would be made regarding H(0): population means are equal? (a) Reject H(0) at the .05 level (b) Fail to reject H(0) at the .01 level (c) Insufficient information is given to answer

Back to this chapter's Contents

Q: In reading a scientific article you encounter the following table: Analysis of Variance ------------------------------------------------------------------------ Source SS df MS F ------------------------------------------------------------------------ Between samples 722.7 4 180.68 15.3** Within samples 473.3 40 11.83 ------------------------------------------------------------------------ Total 1196.0 44 Further reading indicates that all sample sizes are equal. Then we know that the experimenter used (a) 4 samples of size 10. (b) 5 samples of size 10. (c) 4 samples of size 9. (d) 5 samples of size 9. (e) None of these

Back to this chapter's Contents

Q: NOTE: Items i. thru iii. are based on a teaching experiment involving four elementary statistics classes. Below are scores for 24 students who took the same final examination. Statistics Statistics Statistics Statistics Cookbook with Humor Made Useful in Story Form ------------------------------------------------------------------------ 78 51 64 54 78 57 54 61 79 64 61 79 70 75 66 69 83 42 57 69 74 83 71 65 ------------------------------------------------------------------------ 462 372 373 397 Suppose further that calculations yield SS(total) = 800, and SS(between) = 300. i. What is the observed value of the statistic one computes to test H(0): MU(1) = MU(2) = MU(3) = MU(4) against H(1): not all 4 means are equal? (a) 2.5 (b) 5.2 (c) 4.0 (d) 11.1 (e) None of these ii. If ALPHA = .01, then the critical value of the statistic is (a) 3.10 (b) 4.94 (c) 8.66 (d) 26.7 (e) None of these iii. Suppose that the observed value of the statistic in Item i. is 5.6 while the critical value in Item ii. is 7.21. With only this information, which of the following conclusions is most logical? (a) The four populations do not all have the same mean. (b) The four populations have the same mean. (c) The four populations do not all have the same mean. Statistics Cookbook and Statistics in Story Form produce higher means than the other two books. (d) Statistics Cookbook has the highest population mean. (e) The four population means may be different but these samples fail to demonstrate any difference.

Back to this chapter's Contents

Q: An experiment was conducted as a oneway random ANOVA design yielding K sample means, each based on n scores. If the between and within mean squares are represented by S(m)**2 and S(p)**2, respectively, what is the number of degrees of freedom for S(m)**2? a. n - 1 b. K - 1 c. n - K d. (n - 1)(K - 1) e. none of the above

Back to this chapter's Contents

Q: Suppose we have five independent groups (or samples), each of size 9. How many degrees of freedom do we have for the within groups sum of squares in the ANOVA assuming a single factor experiment? 1. 4 2. 41 3. 44 4. 45 5. Not given.

Back to this chapter's Contents

Q: Samples of size 5 are taken from 3 populations and the following analysis of variance table found. Test the hypothesis that the three populations have the same means. Source d.f. M.S. F ------ ---- ---- - Between means 350 Within samples 100 Total a. F = 3.5 so hypothesis is rejected at 5% level b. F = 3.5 so hypothesis is not rejected at 5% level c. F = 3.5 so hypothesis is not rejected at 10% level d. F = 3.5 so hypothesis is rejected at 1% level e. Do not have enough information to perform test.

Back to this chapter's Contents

Q: A fisheries researcher wishes to conclude that there is a difference in mean weights of three species of fish caught in a large lake near Lin- coln, Nebraska. The data are as follows: (Use ALPHA = .05.) SPECIES ------- X Y Z 1.5 1.5 6.0 4.0 SUM(X) = 13 1.0 SUM(Y) = 9 4.5 SUM(Z) = 20.5 4.5 4.5 4.5 3.0 2.0 5.5 ANOVA Table (incomplete) : Source of Variation SS df MS F Between Groups 17.04 2 8.52 Within Groups 14.19 9 1.58 TOTAL 31.23 11 i) The null hypothesis is: a) H(O): BETA = 0 b) H(O): MU = 0 c) H(O): MU(X) = MU(Y) = MU(Z) d) H(O): BETA(X) = BETA(Y)= BETA(Z) ii) The test statistic is: a) t(calc) = 2.52 b) t(calc) = 3.09 c) F(calc) = 1.20 d) F(calc) = 5.40 iii) The critical value is: a) t(.05,9) = 2.262 b) t(.10,9) = 1.833 c) F(.05,2,9) = 4.26 d) F(2.5,2.9) = 5.71 iv) What is your conclusion? a) Reject H(O) because F(calc) > F(crit), (at least 1 pair has different means). b) Reject H(O) because ]t(calc)] > t(crit), (all means are different). c) Fail to reject H(O) because F(calc) < F(crit), (insuffi- cient evidence that means are different). d) Fail to reject H(O) because ]t(calc)] < t(crit), (means are equal).

Back to this chapter's Contents

Q: If the sample means for each of the treatment groups were identical, what would be the observed value of the ANOVA F-ratio? a. 1.0 b. 0.0 c. A value between 0.0 and 1.0 d. A negative value e. Infinite

Back to this chapter's Contents

Q: Assuming a single factor experiment is being analyzed and that the null hypothesis being tested by the F-test in the ANOVA is false, the probability of obtaining a mean-square ratio exceeding that reported in the F table as the 95th percentile is a. greater than .05. b. less than .05. c. equal to .05.

Back to this chapter's Contents

Q: A one-way classification analysis of variance is performed on experi- mental data for which there were 10 subjects in each of two groups. The .95 confidence interval around the difference YBAR(1) - YBAR(2) is -0.10 to 1.5. Which one of the following statements is true? (a) The F ratio obtained in the analysis of variance was less than 4.41 (b) The F ratio obtained in the analysis of variance was greater than 8.28 (c) The true difference MU(1) - MU(2) must lie between -0.10 and 1.50. (d) The best estimate of MU(1) - MU(2) possible from the results is 1.50.

Back to this chapter's Contents

Q: Samples from 3 classes were given an identical math test. The scores are given as follows. Test the hypothesis that the performances of the 3 classes are equal. (ALPHA = .05) Class A Class B Class C ------- ------- ------- 5 10 7 4 2 7 6 1 6 10 5 5 4 2 7 7 7 8 3 8 10 9 2 1 --- --- --- Total 48 37 51 Source of variation SS df MS F Between classes 13.583 2 6.7915 Within classes (error) 220.75 21 10.5 Total 234.333 23

Back to this chapter's Contents

Q: To test the hypothesis that shelf placement influences sales, a marketing researcher has collected data on sales in a random sample of 15 comparable supermarkets with 3 different shelving policies for an identical brand of soup. The data is weekly sales figures (in tens of cans). Perform the appropriate test at the 5% level. If you reject, which shelving policies are different? bottom shelf middle shelf top shelf sales sales sales ------------ ------------ --------- 10 25 10 5 20 10 10 25 20 10 30 20 15 50 40 MEANS FOR SHELF TREATMENT MEAN(SALES) MIDDLE 30 TOP 20 BOTTOM 10 ANALYSIS OF VARIANCE (incomplete) : --------------------- SOURCE OF VARIATION DF SS MEAN SQUARE F(CALC.) SHELF 2 1000.0000 500.00000 EXPERIMENTAL ERROR 12 1200.0000 100.00000 TOTAL 14 2200.0000 HSD FOR ABOVE MEANS IS 16.86 at CONF.LEVEL 95 (Note: HSD means Tukey's Honest Significant Difference.)

Back to this chapter's Contents

Q: A reseaecher assigns each of his interviewers a list of 7 families, drawn randomly from a region, to be interviewed. Each interviewer is instructed to administer a successful parenting scale (SPS) to each parent in his sample. The SPS scores, Y(i), are defined as ranging from 0 (no parenting skills deemed successful) to 100 (successful parenting skills consistently and skillfully applied). An interviewer returns with data for both parents. Use this data to test, using classical analysis of variance, at the 90% level of con- fidence, the hypothesis that "mothers are more likely to be success- ful parents". Mothers Fathers Y(i) Y(j) 68 63 72 48 48 30 54 52 83 55 92 41 87 57 MBAR = 72.00 FBAR = 49.43 ANOVA Table (incomplete) : Source of Variation SS df MS F(calc) Between Groups 1783.143 1 Within Groups 2391.714 12 199.310 Total 4174.857 13

Back to this chapter's Contents

Q: Use one-way analysis of variance, with an F test, to test the hypothesis that "The wealthier a person, the more likely he will be relatively politically conservative," at the 90% level of confidence. Note that, for purposes of research, the researcher operationally defined "wealthy" as those with an annual income of $50,000, while "poor" subjects re- ceived less than $5,000/year. Note, too, that the "political conser- vatism" scale used produced scores of 0 for "extremely liberal", and 100 for "extremely conservative". Sample data: X, Income Category Wealthy Poor +---------+--------+ | Y Y | Y | | 90 | 50 | | 80 | 60 | | 70 | 40 | | 60 | 50 | | 90 | 30 | +---------+--------+ ANOVA Table (incomplete) : Source of variation | SS | df | MS | F(calc.) --------------------+--------+--------+--------+----------- Between groups | 2560 | 1 | | Within groups | 1200 | 8 | | Corrected total | 3760 | 9 | |

Back to this chapter's Contents

Q: A sociologist conducted a study of assertion by having one of her top students, after appropriate training, note the number of assertive acts performed in a day by each of 10 randomly selected coeds, producing the following sample of data, in acts per day: [5,3,10,6,4,9,5,5,7,5]. Another sociologist wonders whether the male and female students at Bedrock indeed differ in assertiveness and, by a similar procedure, gathers the following data for male students, in acts per day: [8,3,5, 8,12,10,7,7,9,7]. Use a one-way analysis of variance to test the hypothesis that male students are more assertive than female students at Bedrock College, at the 90% level of confidence. ANOVA table (incomplete) : Source of variation | SS | df | MS | F(calc.) ---------------------+--------+--------+--------+---------- Between Groups | 14.45| 1 | | Within Groups | 99.30| 18 | | Total | 113.75| 19 | |

Back to this chapter's Contents

Q: Mr. Martin can drive to work along four different routes, and the following are the number of minutes in which he timed him- self on five different occasions for each route: Route 1 Route 2 Route 3 Route 4 ________________________________________ 22 25 26 26 26 27 29 28 25 28 33 27 25 26 30 30 31 29 33 30 ________________________________________ T.(j) = 129 135 151 141 ANOVA Table (incomplete) Source df SS M.Sq. F ratio Routes .3 52.8 Error 16 100.4 Total 19 153.2 Complete the ANOVA Table and test if all routes are equally fast (ALPHA = 5%).

Back to this chapter's Contents

Q: An imaginary study has been conducted on the effects of three brands of laxatives on regularity of TV actresses where each brand was tested by one actress belonging to each of 10 age groups. Results obtained included: F= (brand M.Sq.)/(Error M.Sq.) = 2.1 with 2 and 18 df. a. What hypothesis is tested using this F ratio? b. Interpret these results using a significance level of 5%.

Back to this chapter's Contents

Q: Samples of size 11 are taken from each of 5 populations. Complete the following analysis of variance table: Source S.S. d.f. M.S. F ------ ---- ---- ---- - Between means 1000 a c e Within samples 5000 b d Total 6000 a. a = 4 b = 44 c = 250 d = 113.6 e = 2.2 b. a = 4 b = 44 c = 250 d = 113.6 e = 0.2 c. a = 5 b = 55 c = 200 d = 90.9 e = 0.2 d. a = 5 b = 50 c = 200 d = 100 e = 2.0 e. a = 4 b = 50 c = 250 d = 100 e = 2.5

Back to this chapter's Contents

Q: In a single factor experiment with four levels if the mean square (between)=25, mean square(within)=10, n(1)=n(2)=n(3)=8 and n(4)=10, what is the value of (corrected) total sums of squares? a. 435 b. 786 c. 1221 d. Insufficient information e. Sufficient information but correct value is not given

Back to this chapter's Contents

Q: In the ANOVA for a single factor experiment with four levels all n's equal 5 and YBAR(1)=22, YBAR(2)=24, YBAR(3)=20, and YBAR(4)=18. What is the sum of squares for between groups? a. 25.00 b. 33.33 c. 100.00 d. Cannot be determined without more data

Back to this chapter's Contents

Q: Nine children were randomly split into three groups of three each. In a spelling unit, individuals in one group were criticized each time they misspelled a word. The individuals in another group were praised each time they correctly spelled a word, while the individuals in the third group were neither praised or criticized. At the end of the unit, each child was given ten words to spell with the following results (number correct is given for each child): Praised Neutral Criticized 8 3 9 9 2 10 7 5 7 MEANS FOR RESPONSE TREATMENT MEAN(WORDS CORRECT) CRITICIZED 8.66667 PRAISED 8 NEUTRAL 3.33333 ANALYSIS OF VARIANCE: -------------------- SOURCE OF VARIATION DF SS MEAN SQUARE RESPONSE 2 50.6667 25.33333 EXPERIMENTAL ERROR 6 11.3333 1.88889 TOTAL 8 62.0000 PROBABILITY LEVEL FOR COMPARING MEANS = 0.05 HSD FOR ABOVE MEANS IS 3.444 AT PROB.LEVEL 0.05 Is there any evidence for significant differences among the methods? Consider each of the pairs of methods. Are there significant differ- ences within any of the pairs?

Back to this chapter's Contents

Q: NOTE: Items i. thru iv. are based on the following situation. Each item indicates how a particular sum-of-squares value is calculated. Use the following code in responding to each item. (a) Between (b) Within (c) Total (corrected) (d) Either (a) or (b) (e) None of these i. Squaring the deviations of each score from its sample (or group) mean. ii. Squaring the deviations of each score from the overall mean. iii. Squaring the deviations of each sample mean from the overall mean. iv. Squaring the deviations of one sample mean from all other sample means.

Back to this chapter's Contents

Q: True or False? If False, correct it. In a one-way ANOVA, the between mean square plus the within mean square must equal the total mean square if computations are correct.

Back to this chapter's Contents

Q: A physiologist reports an investigation of potential plant hormones. He reports the following averages for lengths of 20 stem segments treated. Compound X 1.18 Compound Y 1.17 Compound Z 1.15 Control 1.14 The conclusion is that there are no treatment differences. Are you satisfied with this conclusion? Why or why not?

Back to this chapter's Contents

Q: When various methods for comparing means are discussed two methods that are often cited as extremes are - a. LSD for which the Type I error rate is ALPHA per comparison b. Tukey's HSD, for which the Type I error is ALPHA per experiment regardless of the number of comparisons made. If there are many comparisons to be made in an experiment, use of an LSD is much more exposed to Type I error than use of Tukey's method, but use of an LSD is much less exposed to Type II errors. In each of the items below indicate whether or not you would use an LSD to compare means. Explain your choice. a. An experiment involves 10 treatments. If you commit a Type I error, there will be little immediate loss. If you commit a Type II error there will be somewhat greater losses. b. An experiment involves 2 treatments. If you commit a Type I error, there will be appreciable loss. If you commit a Type II error, there will be little loss. c. An experiment involves 20 treatments. If you commit a Type I error, there could be widespread injury (e.g. loss of life). If you commit a Type II error, there will be bothersome losses (loss of investment in developing treatments), but less loss than from Type I error.

Back to this chapter's Contents

Q: A F-ratio signficant at the .05 level was found in a one-factor ANOVA in which there are 3 levels and n=25. A set of simultaneous confidence intervals with confidence coefficient .95 is established around differences between pairs of means using Tukey's HSD method. The following intervals are obtained: YBAR(1) - YBAR(2): (-7.50, -7.00) YBAR(1) - YBAR(3): (8.50, 8.75) YBAR(2) - YBAR(3): (-.80, 1.50) Which of the following pairs of sample means should NOT be judged to be significantly different? (a) YBAR(1) - YBAR(2) (b) YBAR(1) - YBAR(3) (c) YBAR(2) - YBAR(3) (d) None of the above

Back to this chapter's Contents

Return to the list of chapters

Return to Brian Schott @ GSU

A: F(95th percentile, df=4,5) = 5.19

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: True

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: d. the analysis of variance Of the four tests given, the analysis of variance is the only exact test that can be performed. The t-test does not compare more than two means and chi square is not exact.

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: a) no The confidence interval includes zero.

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: a. (7-0) +/- HSD 7 +/- 2 Interval is from 5 to 9 (significant difference) b. (8-7) +/- 2 Interval is from -1 to +3 (not a significant difference)

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: c) The significance level is given by the area to the right of 9.48773 under the appropriate F distribution.

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: (e) Fail to Reject H(O) at ALPHA = .01. Since P > .01 you cannot reject H(O) at ALPHA = .01.

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: a. no b. yes, a variance estimate is needed in order to estimate sample size required when someone else investigates this phenomenon in the future. c. yes, if circumstances that produce no response aren't reported, the reports on record sometimes will give the misleading impression that there generally is a response. d. no, Type I error rate is to be chosen before the investigation and is not to be changed in order to declare significance.

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: False, good statistical procedure would use an F-test and Tukey's test to avoid the problem of increased type I error in the multiple comparisons.

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: Factors that might influence variability in weights include: variability in height, age, level of activity, diet, and random error.

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: False - When data contains less variability than would be expected by chance, it is time to be suspicious of the techniques or procedures used in collecting the data. But business managers are typically con- servative and prefer data with less variability.

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: 1. The estimated difference between Brands X and L in this trial was 3. Since a follow up study has been commissioned, it would seem that a difference at least this small is to be detected. 2. The mean square for experimental error gives us an estimate of the variance and an estimate of the standard deviation would be the square root of that, so SQRT(100) = 10. 3. As explained above, it is 100.

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: c) 40 SUM(X) = 72 XBAR = 72/8 = 9 Corrected Sum of Squares = (0**2) + (-3**2) + (5**2) + (0**2) + (1**2) + (-2**2) + (-1**2) + (0**2) = 9 + 25 + 1 + 4 + 1 = 40

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: False. It also depends on how many independent parameters (or treatment level means) are being estimated.

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: True.

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: b. MS(within)

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: Definition: A sum of squares divided by its degrees of freedom. Example: If an ANOVA table includes: Source df SS Experimental Error 12 3600, the mean square for experimental error is 300 = 3600/12.

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: b) There is a difference in response due to treatments.

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: c. There was more variability between subjects within the same group than there was between groups. This is precisely the null hypothesis of the F-test in this analysis. Since the MS(within) is larger than MS(among), this F-test will have a calculated value less than one. Based on this result we will undoubtedly not reject the null hypothesis.

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: (a) Reject H(0) at the .05 level Completed ANOVA table: Source SS df MS ------ -- -- -- Between 30.5 4 7.625 Within 134.5 95 1.416 Total 165.0 99 F(calc.) = 7.625 / 1.416 = 5.385 F(crit., ALPHA=.05, df=4,95, one-tail) = 2.47 F(crit., ALPHA=.05, df=4,95, one-tail) = 3.59 H(0) would be rejected at both ALPHA levels

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: (d) 5 samples of size 9.

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: i. (c) 4.0 ANOVA Table: Source SS df MS F ------ -- -- -- - Between samples 300 3 100 4.00 Within samples 500 20 25 Total (corrected)800 23 ii. (b) 4.94 F(crit., ALPHA=.01, df=3,20, one-tail) = 4.94 iii. (e) The four population means may be different but these samples fail to demonstrate any difference since F(obs.) < F(crit.).

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: b. K - 1 Total Categories = K Degrees of Freedom = K - 1

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: 5. Not given. df(within) = (total # of observations) - 1 - df(between) = (5*9) - 1 - 4 = 45-5 = 40

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: b. F = 3.5 so hypothesis is continued at 5% level. Completed ANOVA table: Source d.f. M.S. F ------ ---- ---- - Between samples 2 350 3.5 Within samples 12 100 Total 14 F(critical, df = 2, 12, ALPHA = .05) = 3.89 Therefore the hypothesis should not be rejected.

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: i) c) H(O): MU(X) = MU(Y) = MU(Z) ii) d) F(calc) = 5.40 iii) c) F(.05,2,9) = 4.26 iv) a) Reject H(O) because F(calc) > F(crit), (at least 1 pair has different means). ANOVA Table: Source of Variation SS df MS F Between Groups 17.04 2 8.52 5.40 Within Groups 14.19 9 1.58 TOTAL 31.23 11

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: b. 0.0 because Sum of Squares between would be zero and the F-ratio is: F(calc.) = [SS(between)/df(between)]/[SS(within)/df(within)]

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: a. greater than .05. The null hypothesis being tested is: H(O): MU(1) = M(2) = ... = MU(K) where K is the number of experimental groups or samples. But if this is false, then the K population means are not equal and the between groups mean square will be greater than the within groups mean square. So the sampling distribution of this ratio will be centered further to the right than the F-distribution associated with the null hypothesis.

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: (a) The F ratio obtained in the analysis of variance was less than 4.41

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: H(O): MU(1) = MU(2) = MU(3) H(A): MU(1) =/= MU(2) =/= MU(3) F(calculated) = MS for classes/MS for error = 6.7915/10.5 = .646 F(critical, df=2, 21, ALPHA =.05) = 3.47 Since F(calculated) < F(critical) do not reject H(O). Therefore performances of the 3 classes are not found to be different.

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: PROBABILITY LEVEL FOR COMPARING MEANS = .05 F(critical, df=2,12, ALPHA=.05) = 3.88 Therefore, reject the null hypothesis that shelving policy does not influence sales. Based on the HSD given above, it appears that there is a significant difference between the middle shelf and the bottom shelf.

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: n(M) = n(F) = 7 H(O): MU(females) = MU(males) H(A): MU(females) =/= MU(males) ANOVA Table: Source of Variation SS df MS F(calc) Between Groups 1783.143 1 1783.143 8.947 Within Groups 2391.714 12 199.310 Total 4174.857 13 F(crit., ALPHA=.10, one-tail, df=1,12) = 3.18 Since F(calc) > F(crit), H(O) should be rejected. Thus, based on this sample evidence, it appears that the means for mothers and fathers are different and the indication from the sample means is that mothers are more likely to be successful parents.

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: H(0): MU(Wealthy) = MU(Poor) H(A): MU(Wealthy) =/= MU(Poor) ANOVA Table: Source of variation | SS | df | MS | F(calc.) --------------------+--------+--------+--------+----------- Between groups | 2560 | 1 | 2560 | 17.1 Within groups | 1200 | 8 | 150 | Total | 3760 | 9 | | F(crit., df=1,8, ALPHA=.10) = 3.46 Since F(calc.) > F(crit.) reject H(0) and based on the fact that wealthy people have a higher mean income assume that the wealthier a person, the more likely he will be relatively politically conservative.

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: H(0): MU(Males) = MU(Females) H(A): MU(Males) =/= MU(Females) ANOVA table: Source of variation | SS | df | MS | F(calc.) ---------------------+--------+--------+--------+---------- Between Groups | 14.45| 1 | 14.45 | 2.62 Within Groups | 99.30| 18 | 5.52 | Total | 113.75| 19 | | F(crit., df=1,18, ALPHA=.10) = 3.01 Since F(calc.) < F(crit.), do not reject H(0). Therefore, conclude that there is no difference between men and women at Bedrock College.

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: H(0): The means for the routes are all equal. H(A): The means for the routes are not all equal. ANOVA Table Source df SS M.Sq. F ratio Routes .3 52.8 17.6 2.805 Error 16 100.4 6.275 Total 19 153.2 F(ALPHA = .05, df = 3,16) = 3.24 Since the F ratio is less than 3.24, we continue the null hypothesis that all routes are equally fast.

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: a. The null hypothesis tested is that all brand effects equal zero, H(O): MU(1) = MU(2) = MU(3). b. F(calculated) = 2.1 F(critical, df = 2.18, ALPHA = .05) = 3.55 Since the calculated F value is smaller than the critical F value, we do not reject the null hypothesis that all brand effects are equal.

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: e. a = 4 b = 50 c = 250 d = 100 e = 2.5 a = number of treatments - 1 = 5 - 1 = 4 b = (a+1)*(number of replications - 1) = (4+1)*(11-1) = 50 c = 1000/a = 250 d = 5000/b = 100 e = c/d = 250/100 = 2.5

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: e. Sufficient information but correct value is not given Known parts of ANOVA table are: Source of Variation df SS MS ------------------- -- -- -- Between (Treatment) 3 25 Within 30 10 (Corrected) Total 33 Therefore between Sum of Squares is: SS(B) = mean square(between) * df(between) = (25*3) = 75 Within Sum of Squares is: SS(W) = mean square(within) * df(within) = (10*30) = 300 Corrected Total SS is: SS(T) = SS(B) + SS(W) = 75+300 = 375

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: c. 100.00 Overall Mean = (22+24+20+18)/4 = 84/4 = 21 SS(between) = [5*((22-21)**2)]+[5*((24-21)**2)]+[5*((20-21)**2)] +[5*((18-21)**2)] = [5*1]+[5*9]+[5*1]+[5*9] = 5 + 45 + 5 + 45 = 100.00

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: F(calculated) = (25.3/1.9) = 13.40 -------------------------------------------------------------- H(O): No difference between treatments H(A): Not all treatment means are the same Critical Region: F(df=2,6) > 5.14 Reject H(O), and conclude that amount and/or kind of praise makes a difference in spelling ability. Using the HSD and the treatment means given above, we see that "criticizing vs. neutral" are significantly different and "prais- ing vs. neutral" are significantly different.

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: i. (b) Within ii. (c) Total (corrected) iii. (a) Between iv. (e) None of these

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: False, in a one-way ANOVA, the between SS plus the within SS must equal the total SS if computations are correct. SS(TOTAL) = SS(Between) + SS(Within)

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: On the basis of the information given I am not satisfied with this con- clusion. No indication of an estimate for the background variation has been given, nor any indication of a basis for comparing these means, such as an HSD with a specified significance level. Due to the lack of information reported, there is no basis for making a conclusion one way or another.

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: a. I would use an LSD because I want to protect against Type II error, not Type I errors. b. I could use either method in this case because when there are only two treatments, the two methods are identical. c. I would not use an LSD in this case, I would use Tukey's test. Tukey's method protects us from Type I errors when more than one comparison is to be made.

Back to review this question

Look at this question's identification

Back to this chapter's Contents

A: (c) YBAR(2) - YBAR(3)

Back to review this question

Look at this question's identification

Back to this chapter's Contents

Return to the list of chapters

Return to Brian Schott @ GSU

Item is still being reviewed Multiple Choice FDISTRIBUTION PROBDISTRIBUTION PROBABILITY T= 2 Comprehension D= 2 General Education ***Statistical Table Necessary***

Back to this chapter's Contents

Based upon item submitted by R. Pruzek - SUNY at Albany True/False FDISTRIBUTION TESTING FTEST PROBDISTRIBUTION PROBABILITY CONCEPT STATISTICS ANOVA PARAMETRIC T= 2 Comprehension D= 3 General

Back to this chapter's Contents

Based upon item submitted by R. Shavelson - UCLA Multiple Choice ONEWAY/CR TESTING DESIGN ANOVA SIMPLE/COR COMPLETELYRANDOM PARAMETRIC STATISTICS CONCEPT CORRELATION/P T= 5 Comprehension D= 5 General

Back to this chapter's Contents

Based upon item submitted by R. L. Stout & R. M. Paolino - Brown Multiple Choice SIMPLE/CI TESTOFSIGNIFICAN TESTING CONFIDENCEINTERV ESTIMATION CONCEPT STATISTICS T= 5 Comprehension D= 5 General Biological Sciences Sociology

Back to this chapter's Contents

Based upon item submitted by J. Warren - UNH Numerical Answer OTHER/CI LSD I650C CONFIDENCEINTERV ESTIMATION CONCEPT STATISTICS APOSTERIORI MULTIPLECOMPARIS PARAMETRIC T= 5 Computation Comprehension D= 2 General ***Multiple Parts***

Back to this chapter's Contents

Based upon item submitted by D. Kleinbaum - Univ of North Carolina Multiple Choice TESTOFSIGNIFICAN FTEST FDISTRIBUTION CONCEPT STATISTICS ANOVA PARAMETRIC PROBDISTRIBUTION PROBABILITY T= 2 Comprehension D= 5 General

Back to this chapter's Contents

Based upon item submitted by D. Kleinbaum - Univ of North Carolina Multiple Choice TESTOFSIGNIFICAN CONCEPT STATISTICS T= 2 Comprehension D= 3 General

Back to this chapter's Contents

Based upon item submitted by J. Warren - UNH Short Answer HSD TESTOFSIGNIFICAN TYPICALSUMMARY I650C APOSTERIORI MULTIPLECOMPARIS PARAMETRIC STATISTICS CONCEPT T= 5 Comprehension D= 4 General ***Multiple Parts***

Back to this chapter's Contents

Item is still being reviewed True/False TTEST FTEST TYPE1ERROR NEWMANKEULS DUNCANS TUKEYSHSD PARAMETRIC STATISTICS ANOVA CONCEPT APOSTERIORI MULTIPLECOMPARIS T= 5 Comprehension D= 2 General

Back to this chapter's Contents

Item is still being reviewed Short Answer VARIABILITY/P DESCRSTAT/P PARAMETRIC STATISTICS T= 5 Comprehension D= 3 General Biological Sciences

Back to this chapter's Contents

Based upon item submitted by W. J. Hall - Univ. of Rochester True/False VARIABILITY/P RANDOMVARIATION DESCRSTAT/P PARAMETRIC STATISTICS SAMPLINGERROR SAMPLING T= 5 Comprehension D= 4 General

Back to this chapter's Contents

Item is still being reviewed Short Answer VARIANCE STANDARDDEVIATIO OTHER/AN I650C DESCRSTAT/P PARAMETRIC STATISTICS ANOVA T=10 Comprehension D= 4 General ***Multiple Parts***

Back to this chapter's Contents

Item is still being reviewed Multiple Choice VARIANCE SUMSOFSQUARES DESCRSTAT/P PARAMETRIC STATISTICS ANOVA T= 2 Computation D= 2 General

Back to this chapter's Contents

Based upon item submitted by J. Warren - UNH True/False VARIANCE DEGREESOFFREEDOM I650I DESCRSTAT/P PARAMETRIC STATISTICS ANOVA T= 2 Comprehension D= 3 General

Back to this chapter's Contents

Based upon item submitted by J. Warren - UNH True/False VARIANCE DEGREESOFFREEDOM SAMPLESIZE I650I DESCRSTAT/P PARAMETRIC STATISTICS ANOVA SAMPLING T= 2 Comprehension D= 2 General

Back to this chapter's Contents

Item is still being reviewed Multiple Choice VARIANCE/OTHER SUMSOFSQUARES COMPLETELYRANDOM DESCRSTAT/P PARAMETRIC STATISTICS ANOVA T= 2 Comprehension D= 4 General

Back to this chapter's Contents

Based upon item submitted by J. Warren - UNH Definition VARIANCE/OTHER EXPERDESIGN/TERM DESCRSTAT/P PARAMETRIC STATISTICS I650C ANOVA T= 5 Comprehension D= 3 General

Back to this chapter's Contents

Item is still being reviewed Multiple Choice FTEST OTHER/AN ANOVA PARAMETRIC STATISTICS T= 2 Comprehension D= 3 General

Back to this chapter's Contents

Item is still being reviewed Multiple Choice OTHER/AN FTEST ONEWAY/CR ANOVA PARAMETRIC STATISTICS COMPLETELYRANDOM T= 2 Comprehension D= 2 General Education

Back to this chapter's Contents

Item is still being reviewed Multiple Choice OTHER/AN SUMSOFSQUARES ONEWAY/CR ANOVA PARAMETRIC STATISTICS COMPLETELYRANDOM T= 3 Application D= 3 General ***Calculator Necessary***

Back to this chapter's Contents

Item is still being reviewed Multiple Choice ONEWAY/CR OTHER/AN COMPLETELYRANDOM ANOVA PARAMETRIC STATISTICS T= 2 Comprehension Computation D= 3 General

Back to this chapter's Contents

Item is still being reviewed Multiple Choice OTHER/AN ONEWAY/CR FTEST ANOVA PARAMETRIC STATISTICS COMPLETELYRANDOM T= 6 Application Comprehension D= 4 General Education ***Multiple Parts*** ***Statistical Table Necessary***

Back to this chapter's Contents

Based upon item submitted by W. J. Hall - Univ. of Rochester Multiple Choice DEGREESOFFREEDOM ONEWAY/CR DESIGN ANOVA PARAMETRIC STATISTICS COMPLETELYRANDOM CONCEPT T= 2 Comprehension D= 2 General

Back to this chapter's Contents

Item is still being reviewed Multiple Choice DEGREESOFFREEDOM ANOVA PARAMETRIC STATISTICS T= 2 Computation D= 4 General Education

Back to this chapter's Contents

Item is still being reviewed Multiple Choice FTEST ONEWAY/CR DEGREESOFFREEDOM ANOVA PARAMETRIC STATISTICS COMPLETELYRANDOM T= 5 Computation D= 5 General ***Statistical Table Necessary***

Back to this chapter's Contents

Item is still being reviewed Multiple Choice FTEST ONEWAY/CR ANOVA PARAMETRIC STATISTICS COMPLETELYRANDOM T= 5 Application D= 4 Biological Sciences ***Multiple Parts*** ***Statistical Table Necessary***

Back to this chapter's Contents

Item is still being reviewed Multiple Choice FTEST ANOVA PARAMETRIC STATISTICS T= 2 Comprehension D= 3 General Education

Back to this chapter's Contents

Item is still being reviewed Multiple Choice FTEST ANOVA PARAMETRIC STATISTICS T= 2 Comprehension D= 2 General Education

Back to this chapter's Contents

Item is still being reviewed Multiple Choice FTEST ONEWAY/CR ANOVA PARAMETRIC STATISTICS COMPLETELYRANDOM T= 2 Comprehension Computation D= 3 General

Back to this chapter's Contents

Item is still being reviewed Numerical Answer ONEWAY/CR FTEST COMPLETELYRANDOM ANOVA PARAMETRIC STATISTICS T= 5 Computation D= 5 General ***Calculator Necessary*** ***Statistical Table Necessary***

Back to this chapter's Contents

Item is still being reviewed Numerical Answer FTEST ONEWAY/CR LSD ANOVA PARAMETRIC STATISTICS COMPLETELYRANDOM APOSTERIORI MULTIPLECOMPARIS T= 6 Application D= 4 General Business ***Calculator Necessary*** ***Multiple Parts*** ***Statistical Table Necessary***

Back to this chapter's Contents

Item is still being reviewed Numerical Answer ONEWAY/CR FTEST COMPLETELYRANDOM ANOVA PARAMETRIC STATISTICS T= 5 Computation D= 3 General Social Sciences Sociology ***Calculator Necessary*** ***Statistical Table Necessary***

Back to this chapter's Contents

Item is still being reviewed Numerical Answer ONEWAY/CR FTEST COMPLETELYRANDOM ANOVA PARAMETRIC STATISTICS T= 5 Computation D= 3 Social Sciences Sociology General ***Calculator Necessary*** ***Statistical Table Necessary***

Back to this chapter's Contents

Item is still being reviewed Numerical Answer ONEWAY/CR FTEST COMPLETELYRANDOM ANOVA PARAMETRIC STATISTICS T= 5 Computation D= 3 Sociology General ***Calculator Necessary*** ***Statistical Table Necessary***

Back to this chapter's Contents

Based upon item submitted by A. Bugbee - UNH Short Answer ONEWAY/CR FTEST COMPLETELYRANDOM ANOVA PARAMETRIC STATISTICS T= 5 Computation Comprehension Application D= 3 General ***Calculator Necessary*** ***Multiple Parts*** ***Statistical Table Necessary***

Back to this chapter's Contents

Based upon item submitted by J. Warren - UNH Short Answer FTEST FRATIO I650C ANOVA PARAMETRIC STATISTICS APRIORI MULTIPLECOMPARIS T= 5 Computation Comprehension D= 3 General ***Multiple Parts*** ***Statistical Table Necessary***

Back to this chapter's Contents

Item is still being reviewed Multiple Choice ONEWAY/CR FTEST DEGREESOFFREEDOM SUMSOFSQUARES COMPLETELYRANDOM ANOVA PARAMETRIC STATISTICS T= 5 Computation D= 5 General

Back to this chapter's Contents

Item is still being reviewed Multiple Choice SUMSOFSQUARES ONEWAY/CR ANOVA PARAMETRIC STATISTICS COMPLETELYRANDOM T= 5 Comprehension Computation D= 5 General Education

Back to this chapter's Contents

Item is still being reviewed Multiple Choice SUMSOFSQUARES ONEWAY/CR ANOVA PARAMETRIC STATISTICS COMPLETELYRANDOM T= 5 Comprehension Computation D= 6 General Education

Back to this chapter's Contents

Based upon item submitted by D. Zahn - Florida State U. Essay ONEWAY/CR HSD COMPLETELYRANDOM ANOVA PARAMETRIC STATISTICS APOSTERIORI MULTIPLECOMPARIS T= 5 Application D= 5 General Education ***Statistical Table Necessary***

Back to this chapter's Contents

Item is still being reviewed Multiple Choice SUMSOFSQUARES ANOVA PARAMETRIC STATISTICS T= 5 Comprehension D= 3 General ***Multiple Parts***

Back to this chapter's Contents

Based upon item submitted by R. Pruzek - SUNY at Albany True/False SUMSOFSQUARES ANOVA PARAMETRIC STATISTICS T= 2 Comprehension D= 2 General

Back to this chapter's Contents

Item is still being reviewed Short Answer MULTIPLECOMPARIS TESTOFSIGNIFICAN I650C PARAMETRIC STATISTICS CONCEPT T= 5 Comprehension D= 3 General Biological Sciences

Back to this chapter's Contents

Based upon item submitted by J. Warren - UNH Short Answer HSD TUKEYSHSD TYPE1ERROR TYPE2ERROR I650C APOSTERIORI MULTIPLECOMPARIS PARAMETRIC STATISTICS CONCEPT T=10 Comprehension D= 4 General ***Multiple Parts***

Back to this chapter's Contents

Item is still being reviewed Multiple Choice TUKEYSHSD APOSTERIORI MULTIPLECOMPARIS PARAMETRIC STATISTICS T= 2 Computation D= 3 General

Back to this chapter's Contents

Return to the list of chapters

Return to Brian Schott @ GSU