Stats 611 Final Study Set

Term
1 / 24
Which measure of central tendency is the most stable across samples?

a. Mean
b. Mode
c. Standard Deviation
d. Median
Click the card to flip 👆
Terms in this set (24)
Which measure of central tendency is the most stable across samples?

a. Mean
b. Mode
c. Standard Deviation
d. Median
a. Mean
How does the formula for standard deviation differ when calculating it for a population versus a sample? (Note: You don't need to report the whole formula - just identify what is different and generally why (without details).
The difference between the standard deviation (SD) formula for a population versus a sample is that a sample SD formula will have n-1 in the denominator whereas population will have n in the denominator because it is a biased statistic and subtracting 1 (dividing by a small number) will make our point estimate larger.
STILL EDITING//Given the SPSS output below, what is the variance for 'objectn'?

[insert table]
a. 7.071
b. 156.24
c. 0.307
d. 49.99
d. 49.99
A z-score is zero when

a. its raw score is zero
b. its raw score is equal to the mean of its distribution
c. its distribution is normal
d. this is a trick question; you can't get a z score of zero
b. its raw score is equal to the mean of its distribution
The assumption of homogeneity of variance is met when:

a. The variance in one group is twice as big as that of a different group
b. The variance across groups is proportional to the means of those groups
c. Variances in different groups are approximately equal
d. The variance is the same as the inter-quartile range
c. Variances in different groups are approximately equal
Suppose you are told that a list of the weights of 25 persons (measured in pounds) follows the normal curve and that the range of the numbers in the list is 48 pounds. Which of the following is the most likely to be the standard deviation of the numbers in the list?

b. Justify your answer to the previous question in one to two sentences.
8 pounds

b. Under a normal curve with a small sample it is unlikely that anybody would be more than 3 SD away or that everybody would be within 1 SD.

One way to think about the problem is for you to draw a normal curve and put the low point and the high point to correspond to the range with the midpoint corresponding to the mean. Then look at it and think about standard deviations.
What is an acceptable way to resolve outliers in one's data?

a. Replace with less extreme value
b. Perform a nonlinear transformation of data
c. Exclude them from any statistical analyses
d. Convert them to z-scores
d. Convert them to z-scores
What word should be in the numerator?

?????/ Error

Briefly, say something about these two concepts and how they relate to each other and statistics.
Effect is what we are interested in (systematic variation, signal). We want to maximize it. Error is that which we try to minimize (unsystematic variation, noise). Most statistics boil down to this. Big effect good, big error bad.
Which of the following transformations would preserve the original shape of a distribution of scores?

a. Calculating the inverse of each score (ex: 1/X)
b. Taking the square root of each score
c. Adding 3 to each score
d. Finding the logarithm to each score
c. Adding 3 to each score
If your raw score on the SAT's (M=1000, s = 100) was converted to a z-score of 2.5, then your raw score was _____ .

a. 1205
b. 1250
c. 1300
d. 1350
b. 1250

z-score formula = (x - μ) / σ
All of the following are measures of variability EXCEPT ______. a. Standard deviation b. Mean c. Variance d. Rangeb. MeanTrue/False: More items on a scale will decrease Cronbach's AlphaFalseTrue/False: Validity is accuracy and reliability is consistency.TrueIn Classical Test Theory, x = t + e. What does the "x" represent? a. the mean of values representing the latent construct b. the observed value c. the assumed value of the latent construct d. the error associated with measuring the latent constructb. the observed valueTo ask if survey items represent all relative components of a construct would be measuring: a. Content validity b. Face Validity c. Test-retest reliability d. Criterion validitya. Content validityTrue/False: We always have sampling error when taking a sample from a populationTrueThe Central Limit Theorem states that: a. Large sampling distributions will take the shape of a normal distribution b. SD of sampling distribution=SD of the sample/square root of N c. The sampling distribution can be normal even if one's variable in a sample is not normal d. All of the aboved. All of the aboveIn the formula for variance, what does the numerator represent? What does that mean in English?The numerator refers to the sum of squares or sum of squared deviations. It is just how far each individual score is from the mean summed across all individuals.True/False: The standard deviation of the DSM is called the standard error of the meanTrueIf we fail to reject the H0 based on the available data when in fact H0 is really false, we will be committing a a. Type I error b. Type II error c. Null Hypothesis d. Alternative Hypothesis e. None of the aboveb. Type II errorWhich of the following provides you with a measure of how well your sample approximates the population? a. Reliability b. Population c. Sampling error d. Hypothesisc. Sampling errorA null hypothesis: a. states that the experimental treatment will have an effect b. is rarely used in experiments c. predicts that the experimental treatment will have no effect d. none of the abovec. predicts that the experimental treatment will have no effectJeff continues to say that all statistics boil down to _______over _______. a. error; effect b. mean; reliability c. noise; signal d. effect; errord. effect; errorTrue/False: If a result is statistically significant, it cannot be due to chance.False