Coefficient of Determination and Nonparametric Tests Statistics Questions
Please answer the attached 41 statistics questions. No need to show work. Just select the correct answer choice. 1.When the regression line is written in standard form (using z scores), the slope is called zr.True or False? 2. The coefficient of determination is given by SSExplained/SST.True or false? 3. Nonparametric tests are useful for data that are measured only at the nominal or ordinal level of measurement.True or false? 4. The χ2 distribution is a positively skewed distribution.True or false ?5. The χ2 goodness of fit test and the χ2 test of independence are useful only if you have interval/ratio data.True or False? 6. Dependent samples, where each is measured on one dichotomous variable, can be analyzed using the McNemar test for significance of change.True or false? 7. Parametric tests are usually more powerful than their nonparametric analogs.Group of answer choicesTrue or False ? 8. A value of Y as predicted from the regression line is symbolized by .True or false? 9. A scatter diagram plots which of the following? Group of answer choicesthe distribution of the dependent variablethe distribution of the independent variableresiduals on one axis and the independent variable on the otherone variable on the X axis and another on the Y axis10. If it is impossible to determine whether the points in a scatter diagram tend from lower right to upper left or from upper right to lower left, the correlation is probablyGroup of answer choicesnegativepositivezero or close to zero11. Which of the following values for a correlation coefficient would indicate the strongest degree of relationship?Group of answer choices-.69+.59 c. +.03 d. -.3512. If ΣzXzY = 4 and n = 9, r isGroup of answer choicesa. +.50b. -.50c. 0d. impossible to determine from the information provided13. _ is the appropriate correlation coefficient for interval/ratio data; ____ is the appropriate correlation coefficient for ordinal data.Group of answer choicesSpearman's rS / Pearson's rSpearman's rS / no statisticPearson's r / Spearman's rSPearson's r / no statistic14. Exhibit 16-1Suppose you are interested in the relationship between number of piano lessons and skill as a piano player. You obtain reports about the number of lessons 50 children have had, and then have judges rate the skill of the piano players on a scale from 1 to 100 (100 being "excellent"). Then you compute a correlation coefficient for these data.Refer to Exhibit 16-1. You would expect that the correlation coefficient would probably beGroup of answer choicesonenegativepositivezero15. The appropriate standard score formula for Pearson's product-moment correlation coefficient for a sample isGroup of answer choicesρ = r = ρ = r = 16. A correlation coefficient ranges from ____ to ____.Group of answer choicesThe correlation coefficient can take any value and is not limited by any range.0 to 1 c. 0 to 100 d. -1 to 117. If you superimpose a tic-tac-toe diagram on a scatterplot, a point in the upper left corner of the tic-tac-toe diagram tends to make the correlation coefficient ____ because ____.Group of answer choicesnegative, zX and zY are both negativepositive, zX and zY are both negativenegative, zX is negative but zY is positivepositive, zX is negative but zY is positive18. The values of Y as predicted from the regression line are symbolized by?19. The error of prediction for the ith individual is given byGroup of answer choicesei = Yi = ei = ei = 20. The slope of a straight line is defined asGroup of answer choicesthe change in Y divided by the change in Xthe value of Y when X = 0the value of X when Y = 0the change in X divided by the change in Y21. The intercept of a straight line is defined asGroup of answer choicesthe value of Y when X = 0the change in Y divided by the change in Xthe change in X divided by the change in Ythe value of X when Y = 022. When the regression line is written in standard form (using z scores), the intercept isGroup of answer choiceszarzb023. Assume that = 29.6 and sY = 5.3. In which case is the standard error of estimate smallest?Group of answer choicesr = .7; sX = 4r = .9; sX = 12r = .4; = 100r = .1; = 1024. The coefficient of determination isGroup of answer choicesalways larger than sYanother name for the Pearson ralways smaller than sYequal to r225. The coefficient of determination is equal toGroup of answer choicesSSExplained/SSTrSSError/SST(SSExplained + SSError)/SST26. If a car has ten independent steering wheels with ten independent drivers, and each contributes equally to the direction that the car takes, the correlation between any one steering wheel and the direction the car takes is approximatelyGroup of answer choices0.1.9.327. Assume that IQ has mean 100 and standard deviation 15 for both boys and girls, and that the correlation between brother's and sister's IQ is .6. Suppose you know that John's IQ is 110. What should you predict his sister's IQ to be?Group of answer choices1151101009410628. What is ?Group of answer choicesThe equation for predicting variable X from variable YThe equation for the regression lineThe equation for the true scoreThe equation to decide why to wear a hat29. What two pieces of information are needed when drawing a regression line?Group of answer choicesThe value of X and the value of aThe standard error of the estimate and the correlation coefficientThe slope and the interceptRaw score and true score30. When are you most likely to see regression to the mean?Group of answer choicesWhen you select a sample based on extremely low scoresWhen the correlation coefficient is smallWhen you select a sample based on extremely high scoresAll of the above31. "nonparametric""parametric"DesignAt least nominalAt least ordinalNormalOne sample<Problem 1>--One samplet or z testTwo independentχ2 test of<Problem 2>Two independ.samplesindependencesample t testTwo dependent<Problem 3>Wilcoxon matchedTwo dependentsamplespairs signed-sample t testranks testk independentχ2 test of<Problem 4>One-waysamplesindependenceANOVARefer to Exhibit 18-1. The table shows comparisons between parametric and non parametric tests. Which test should replace the one marked <Problem 1>?Group of answer choicesMann-Whitney U-testGoodness of fit testKruskal-Wallis H-testStevens-Johnson S-testMcNemar test for significance of change32. If we had nominal data and our null hypothesis was that the shape of a distribution did not depend on the value of another variable we would use theGroup of answer choicesχ2 goodness of fit testWilcoxon matched-pairs signed-rank testχ2 test of independenceMann-Whitney U test33. Which of the following is true?Group of answer choicesThe χ2 statistic is a symmetric distribution.The χ2 statistic, like z, has no degrees of freedom.One must specify, as one does when using F, two sets of degrees of freedom when using the χ2 distribution.The χ2 statistic approaches a normal distribution when the degrees of freedom is large.34. Exhibit 18-2Suppose I build a card shuffler, and I wish to know whether it is effective. I take a standard bridge deck (52 cards in 4 suits: 13 Clubs, 13 Diamonds, 13 Hearts, and 13 Spades), use my shuffler on it, and draw a card, noting its suit. I repeat this process 100 times, and obtain the following sample distribution: 23 Clubs, 25 Diamonds, 26 Hearts, and 26 Spades.Refer to Exhibit 18-2. The proper test to use isGroup of answer choicesMann-Whitney U-testgoodness of fit testWilcoxon testKruskal-Wallis H-test35. Exhibit 18-2Suppose I build a card shuffler, and I wish to know whether it is effective. I take a standard bridge deck (52 cards in 4 suits: 13 Clubs, 13 Diamonds, 13 Hearts, and 13 Spades), use my shuffler on it, and draw a card, noting its suit. I repeat this process 100 times, and obtain the following sample distribution: 23 Clubs, 25 Diamonds, 26 Hearts, and 26 Spades.Refer to Exhibit 18-2. Which of the following is true?Group of answer choicesThe appropriate statistic is FThe number of degrees of freedom is 1My shuffler is biased because the sample distribution should have been 25, 25, 25, 25The number of degrees of freedom is 336. Exhibit 18-3Observed values of two variables:Variable IValue 1Value 2Row TotalValue 11020 30Variable IIValue 23040 70Column Total4060100Refer to Exhibit 18-3. If we had a 4 × 4 contingency table, degrees of freedom for the x2 test of independence would beGroup of answer choices1512916837. Exhibit 18-3Observed values of two variables:Variable IValue 1Value 2Row TotalValue 11020 30Variable IIValue 23040 70Column Total4060100Refer to Exhibit 18-3. In the contingency table above, what is the expected frequency of the upper left cell?Group of answer choices4010123038. Suppose we use a 9-point rating scale, and rate four subjects as shown in the table below. Then we convert those ratings to ranks so that the lowest rating is assigned a rank of 1.SubjectRatingGeorge4Sam2Oscar5Bill4Jack8Oscar's rank isGroup of answer choices3415239. When choosing between parametric and nonparametric tests, which of the following is true?Group of answer choicesYou may always perform a nonparametric test even if a parametric test is also appropriate.The larger the sample size, the less critical are violations of assumptions about normality.Analysis of variance has been shown to be robust to violations of the assumption of normality.All of the above are true.Parametric tests are usually more powerful than nonparametric tests.40. There are two kinds of nonparametric tests, those ____ and those ____.Group of answer choicesbased on inferences, based on statisticsbased on parameters, based on pluralismbased on order, not based on orderbased on statistics, based on inferences41. If we had nominal data and our null hypothesis was that the sampled data came from a specified distribution we would use theGroup of answer choicesχ2 test of independenceWilcoxon matched-pairs signed-rank testχ2 goodness of fit testMann-Whitney U test