Cards (43)

  • reaction time and skill as a driver  would be most likely to show a negative correlation.
  • In the "world of real data," we are least likely to find values of  that are very close to 1.00..
  • If  r=.50 and SySx = 90 then the covariance is equal to 45.
  • The magnitude of the covariance is influenced by a variable's metric  is the principal weakness of the covariance.
  • If r = -1.00, then values of Y can be predicted from values of X without error.
  • Fifty students take a 100-item true-false test. Every student attempts every item. For each student, let X be the number of questions answered correctly and Y be the number not answered correctly. We would expect  ٰr-superscript xy to be -1.00.
  • It is possible to compute a correlation coefficient if we have  two sets of scores for the same group of individuals.
  • A causal relationship between X and Y can be inferred only on grounds that go beyond the simple showing of a degree of relationship.
  • In a group of children ages 7 to 13, strength grip was correlated with number of correct answers on an arithmetic test. We would expect the correlation to be moderately positive.
  • Fundamentally, a causal relationship between two variables results in a degree of association between them.
  • In one study of auto drivers, it was found that a lower frequency of accidents was associated with more years of experience and with greater age of the driver. In explaining this finding, it might be that experience is the important factor in having fewer accidents, and age is not,  both age and experience make a contribution to infrequency of accidents, or that  age is the important factor, and experience is not.
  • When a curved line is the line of best fit to the points in a scatterplot, Pearson r  will describe how well the points hug the best fitting straight line.
  • f Pearson  is calculated for data that are curvilinearly related, it will underestimate the degree of association.
  • In a given group, the correlation between height measured in feet and weight measured in pounds is +.68, neither Height is expressed in centimeters. or Weight is expressed in kilograms would alter the value of  r.
  • In working one correlation problem, X scores are recorded to the first decimal (e.g., 12.2, 13.7, etc.). For convenience, the decimal is ignored (e.g., scores are treated as though they were 122, 137, etc.). If the obtained value of  is +.30, the correct value will be +.30.
  • Among a group of children, the correlation between test score in a science course and test score in an English course is +.45. It is learned that each science test score is 5 points too high, so each score is corrected and  r recomputed. We expect that its value will be unchanged.
  • The correlation coefficient is obtained between academic aptitude test score and academic achievement (a) among students in general and (b) among honor students. Other things being equal, we expect the first to be higher.
  • We would expect the correlation between height and weight for the Woodside High basketball team to be lower than the correlation for the entire student body.
  • A group of highly creative men and women are studied to determine the degree of relationship between creativity ratings and IQ scores. For this group we would expect the correlation to be low.
  • The  between job aptitude scores and job success rating sis computed to be +.29. Tthe best guess as to the value for  had all, rather than just the best qualified, applicants been hired would be moderate.
  • In a study concerned with the relationship between X and Y, it is found that 9% of the variance in Y is associated with X. Thus r-superscript xy   must have been .30.
  • An investigator obtains r=+.4  between peer ratings of sociability and scores on the Wilson Sociability Inventory, for which X-bar=50, S-10.  The amount of sociability inventory variance associated with variation in peer ratings is 16.
  • In one study, a correlation of -.49 is found between the number of hours of TV watched per week and high school GPA. According to this study, 24% of the GPA variance is associated with TV watching.
  • Fifty percent of Y variance is associated with variation in X when r  is approximately .7.
  • One way of interpreting the degree of association between two variables involves first doing something to the correlation coefficient. It involves squaring it.
  • The principal weakness of the covariance is the magnitude of the covariance is influenced by a variable's metric.
  • If r = -1.00, values of Y can be predicted from values of X without error.
  • Fifty students take a 100-item true-false test. Every student attempts every item. For each student, let X be the number of questions answered correctly and Y be the number not answered correctly. We would expect   rₓy to be-1.00.
  • It is possible to compute a correlation coefficient if we have  two sets of scores for the same group of individuals.
  • A causal relationship between X and Y can be inferred only on grounds that go beyond the simple showing of a degree of relationship.
  • In a group of children ages 7 to 13, strength grip was correlated with number of correct answers on an arithmetic test. We would expect the correlation to be moderately positive.
  • In one study of auto drivers, it was found that a lower frequency of accidents was associated with more years of experience and with greater age of the driver. In explaining this finding, it might be that experience is the important factor in having fewer accidents, and age is not, or age is the important factor, and experience is not, or both age and experience make a contribution to infrequency of accidents.
  • When a curved line is the line of best fit to the points in a scatterplot, Pearson r  will not describe how well the points hug the curved line, how well the points hug the best fitting straight line, or how well the points hug a line intermediate between the curved line and the straight line.
  • If Pearson  is calculated for data that are curvilinearly related, it will  underestimate the degree of association.
  • In a given group, the correlation between height measured in feet and weight measured in pounds is +.68. What would alter the value of r? No change.
  • In working one correlation problem, X scores are recorded to the first decimal (e.g., 12.2, 13.7, etc.). For convenience, the decimal is ignored (e.g., scores are treated as though they were 122, 137, etc.). If the obtained value of r is +.30, what will the correct value be? +.30.
  • We would expect the correlation between height and weight for the Woodside High basketball team to be lower than the correlation for the entire student body.
  • A group of highly creative men and women are studied to determine the degree of relationship between creativity ratings and IQ scores. For this group we would expect the correlation to be low.
  • The r between job aptitude scores and job success rating sis computed to be +.29. Which of the following is the best guess as to the value for  r had all, rather than just the best qualified, applicants been hired? +.41.
  • In a study concerned with the relationship between X and Y, it is found that 9% of the variance in Y is associated with X. Thus rxy must have been .30.