The statement that a person who scores 120 has twice as much of the trait being measured as someone who scores 60 is appropriate for:
a variable measured on an interval scale
a variable measured on a ratio scale
any continuous variable
any test whose scores are normally distributed
Question 2 0.5 pts
Which of these is a discrete measure?
How many points you earn on a test
How many inches you grew in a year
How many term papers are due this semester
How many hours you study for an exam
Question 3 0.5 pts
It is not possible to report the average score for data measured on what type of scale?
nominal
ordinal
interval
ratio
Question 4 0.5 pts
Jane’s percentile rank on a science exam is 78. This means that:
Jane answered 78% of the questions correctly on the exam
Jane’s score was equal to or greater than 78% of the other students’ scores
Jane achieved a score of 78% on the exam
Jane scored the same as 78% of the other students
Question 5 0.5 pts
Rudy scores 83% on a test. From that, we know that:
Rudy’s percentile rank is 83%
83% of the people who took the test scored the same or lower than Rudy
Rudy scored above average
Out of 100 possible points, Rudy scored 83 pointsQuestion 6 0.5 pts
Here is a set of scores: 3,6,2,9,4,7,5,7,7,5. If the scores were entered into a frequency table, what would be the cumulative frequency?
7
9
10
55
Question 7 0.5 pts
Your data consist of the average family size for all families within each of six different income levels. What type of visual display would best convey the data?
histogram
line graph
bar chart
frequency table
Question 8 0.5 pts
You are constructing a histogram for scores that range from 70 to 100 in whole points. Frequencies range from 3 to 10; that is, every whole-point score between 60 and 100 occurs at least 3 times, and at least one score occurs 10 times. Which of these would probably be the best range and size for the score intervals along the X-axis?
1-point intervals from 70 and 100
1-point intervals from 0 and 100
5-point intervals from 70 and 100
10-point intervals from 70 and 100
Question 9 10 pts
You are looking at a frequency table for a large number of scores. Without doing any further calculating, which measure of central tendency can you immediately report?
mode
median
mean
none of these
Question 10 10 pts
In a set of raw scores for a teacher-constructed test, what will be the shape of the distribution around the mean?
normal
bimodal
skewed
it could be distributed in any of these ways
Question 11 0.5 pts
From the standard deviation, we know something about the:
general location of most scores within a scores distribution
spread of scores within the scores distribution
shape of the scores distribution
frequency at each score value
Question 12 0.5 pts
You select a sample of 50 scores from a population of 2,000 scores. You compute the range and standard deviation on the sample of 50 scores. You then select another sample of 50 scores from the same population. What measure of dispersion is likely to vary most between your first and second samples?
the range
the standard deviation
they will both vary by the same amount
there is no way to know which one will likely vary most
Question 13 0.5 pts
Here is a set of scores: 5, 5, 8, 14, 22, 22. If the set of scores is changed to 6, 6, 8, 14, 21, 21, how will these changes affect the standard deviation?
It will decrease the standard deviation
It will increase the standard deviation
It will have no effect on the standard deviation
There is not enough information to tell
Question 14 0.5 pts
The probability of rolling a 3 on a die is .17 over the long term (under an infinite number of rolls). What is the probability of rolling a 3 on a single roll?
.00
.17
.83
1.00
Question 15 0.5 pts
The probability of drawing a heart from a deck of cards is .25 over the long term (under an infinite number of draws). What is the probability of getting a heart on a single draw?
.00
.25
.75
1.00
Question 16 0.5 pts
A market researcher obtains a list of all streets in a town. She randomly samples 10 street names from the list, and then administers survey questions to every family living on those 10 streets. What type of sampling is this?
simple random
stratified random
cluster
convenience
Question 17 0.5 pts
In an experiment to determine the effect of exercise on weight-loss among postmenopausal women who are employed outside the home, what is the DEPENDENT variable?
exercise
menopausal status
employment status
weight loss
Question 18 0.5 pts
“Dogs that receive regular veterinary care live longer than those who do not receive such care.” What type of hypothesis is this?
null
directional
non-directional
double-barreled
Question 19 0.5 pts
“There will be no difference in the activity level of hyperactive children given medication and those not given medication.” What type of hypothesis is this?
null
directional
non-directional
double-barreled
Question 20 0.5 pts
A researcher plans to conduct a one-tailed test of her hypothesis at the .05 error level. Just before running the experiment, she decides that a two-tailed test is more appropriate. If she does not adjust the error level, how much error will her new study permit in a single tail?
.10
.05
.025
There is not enough information to tell
Question 21 0.5 pts
Professor Stringent conducts his study at a .01 error level. Professor Lax conducts his study at a .10 error level. Whose study is the better study?
Professor Stringent’s
Professor Lax’s
Both professors’ studies are equally worthy
There is not enough information to tell
Question 22 0.5 pts
The probability of a score falling at or below a particular z score is .8734. If you reject the null hypothesis, what is your Type 1 error level?
87.34%
12.66%
0.13%
There is not enough information to tell
Question 23 0.5 pts
What is the probability of a Type 1 error if you reject the null hypothesis based on a z score of .50?
0%
50%
100%
It depends on the individual values of the raw scores
Question 24 0.5 pts
Assume that a population of scores is normally distributed with a mean of 50 and a standard deviation of 10. For samples of size 25, we would expect 95% of the sample means (not of the raw scores) to fall between what two values?
30 and 70
40 and 55
46 and 54
51 and 59
Question 25 0.5 pts
A population has a mean of 300 and a standard deviation of 25. You draw a random sample from that population. What is the sample’s most likely mean?
0
25
300
There is not enough information to tell
Question 26 0.5 pts
If the population mean is 50 and the standard deviation is 5, what is the normal deviate Z for a sample of 20 people with a mean of 46?
-4.46
-3.57
1.12
-0.25
Question 27 0.5 pts
A two-tailed t test study for a single sample of 28 participants yields a t of -2.35. Look up this value in a t-table. Given the fixed number of participants, what is the lowest tabled alpha level at which this t is statistically significant?
1%
2%
5%
10%
Question 28 0.5 pts
In a one-sample t test, if the observed sample mean turns out to be one that would rarely occur when the null hypothesis is true, what should the researcher do?
reject the null hypothesis
retain the null hypothesis
repeat the test until we get a more probable sample mean
change the level of significance (Type I error)
Question 29 0.5 pts
A research report says that t (63) = 1.99; p = .03. From that information, can you reject the null hypothesis with 95% confidence?
yes
no
it depends on the sample size
it depends on the size of the Type 1 error
Question 30 0.5 pts
Which of these confidence intervals would include the widest range of scores?
68%
95%
99%
All confidence intervals are a standard range
Question 31 0.5 pts
What is the mean of any sampling distribution of the difference between the means?
0
The same as the difference between the two sample means in the study
The square root of the combined sample size of the two samples in the study
It depends on the value of the raw scores in the two sampling distributions
Question 32 0.5 pts
For a very small sample size, you should expect the size of the standard error of the difference between the means to be:
larger than usual
about the same size as usual
smaller than usual
almost zero
Question 33 0.5 pts
In a two-sample t test, if the observed difference between the sample means turns out to be one that could easily occur when the null hypothesis is true, what should we do?
Reject the null hypothesis
Retain the null hypothesis
Repeat the test until we get a less probable difference between the means
Change the level of significance (Type 1 error)
Question 34 0.5 pts
Which of these would be a related samples study?
1. Amount of sugar added to coffee by men versus women
2. Running speed for soccer players pre-season versus post-season
3. Number of hours the members of a class study for a weekly quiz versus a unit exam
1 and 2 only
1 and 3 only
2 and 3 only
1, 2 and 3
Question 35 0.5 pts
The hypothesis testing procedure for comparing means of RELATED samples would be appropriate for comparing:
differences in the religious beliefs of best friends
the distance that 12-year-olds versus 14-year-olds can kick a soccer ball
final grades in a course taught by Professor Jones compared with final grades in a course taught by Professor Smith
birth weights of babies born to drug-addicted mothers and birth weights of babies born to mothers who are not drug-addicted
Question 36 0.5 pts
The hypothesis testing procedures for comparing means of INDEPENDENT samples would be appropriate for comparing:
IQ scores of children attending a public elementary school as they progress from grade to grade
husbands’ and wives’ attitudes toward higher education
male and female problem-solving skills (number of moves toward the correct solution) on a spatial puzzle task
software “usefulness” ratings by users who try both Software A and Software B
Question 37 10 pts
“Children whose parents smoke in their presence are more likely to develop asthma than children whose parents do not smoke in their presence.” How many tails should be used when looking up the t test statistic for this study?
one tail
two tails
either one tail or two tails (it doesn’t matter)
there is not enough information to tell
Question 38 0.5 pts
When subjects who are treated in the SAME way differ in performance, what accounts for their measured difference?
between-groups variation
within-groups variation
total variation
treatment variation
Question 39 0.5 pts
When subjects who are treated in the SAME way differ in performance, what accounts for their measured difference?
random error
the treatment
the IV (independent variable)
this cannot happen
Question 40 0.5 pts
“One way” means that:
there is only one correct way to calculate the test statistic
the results can be in only one direction
there is only one sample (group) in the study
there is only one IV (independent variable) in the study
Question 41 0.5 pts
The appropriate statistic to use when testing the hypothesis for a study with three treatment groups is a:
one-sample t test
two-sample t test
ANOVA F test
Pearson’s correlation coefficient
Question 42 0.5 pts
You have conducted an analysis of variance and found no statistical significance in the F. Should you then calculate a post hoc test?
Yes, to see where the hidden significance might lie
Yes, to locate the confounding variables that might have masked the effect
Yes, to validate your non-significant F
No, because there is there is no statistical significance to be found
Question 43 0.5 pts
You conduct an ANOVA at the .05 error level and find a significant F. You conduct a Tukey HSD post hoc test on the same data at the same .05 error level and find no significant difference between the particular groups. From these results, you know that:
the sample size in each group was not sufficient to pick up the effect detected by the F test
the variation within each group masked the variation between the groups
you should have conducted your Tukey HSD at the .01 error level
you have made a calculation error in either the F test or the Tukey HSD
Question 44 0.5 pts
Your chi-square study has 4 categories or conditions of the first variable and 4 levels or conditions of the second variable. How many degrees of freedom are in this study?
7
8
9
16
Question 45 0.5 pts
You are testing the hypothesis that the average salary of college graduates 1 year out of college is related to the graduates’ genders and to the graduates’ fields of study. What is the appropriate analytic technique for this study?
two-sample t-test
analysis of variance
chi-square goodness of fit
chi-square test of independence
Question 46 0.5 pts
You are testing the hypothesis that the grade point averages differs between college athletes and non-athletes. What is the appropriate analytic technique for this study?
two-sample t-test
Pearson correlation coefficient
chi-square goodness of fit
chi-square test of independence
Question 47 0.5 pts
In a correlational study, we seek to determine:
if the independent variable causes an outcome in the dependent variable
which variable is the cause and which variable is the effect
the direction of causation
how well one variable predicts another variable
Question 48 0.5 pts
In a scatterplot, the data for a negative correlation will graph in what direction?
from the bottom left to the upper right
from the upper left to the bottom right
parallel to the horizontal X-axis
parallel to the vertical Y-axis
Question 49 0.5 pts
In a scatter diagram, if one of the points does not fall on the straight line of best fit to the data points, then r cannot be:
0
+1.00 or -1.00
positive
negative
Question 50 0.5 pts
Which one of the following would most likely show a negative correlation?
verbal aptitude and number of books read per year
body weight and age at which puberty was reached
amount of alcohol ingested and driving ability
educational level and income
Question 51 0.5 pts
The usefulness of a screening test for selecting college students for inclusion in an Honors program implies what type of relationship between the screening test and college grade point average (GPA)?
positive
negative
curvilinear
perfect
Question 52 0.5 pts
Which of these correlations is the strongest?
-0.75
0.00
+0.50
+0.67
Question 53 0.5 pts
Assume that, for variables X and Y, the statistics are as follows:
Variable X Variable Y
Mean = 50 Mean = 100
Stand. Dev. = 10 Stand. Dev. = 20
rxy = +1.00
What would be the predicted score on variable Y for a person who scores 40 on variable X? (Note: This can be answered without any formulas or calculations)
40
80
100
120
Question 54 0.5 pts
If r = −1.00, which one of the following will be true?
Scores on X will be of no use in predicting scores on Y.
Scores on X can be perfectly predicted from scores on Y.
The scores on X and Y are both negative.
The relationship becomes difficult to interpret.
Question 55 0.5 pts
For the regression equation Y’ = bX + a, what is the value of the criterion for someone who scores 60 on the predictor, where the correlation is .75, and where the prediction line’s Y-intercept is 5?
45
48.75
50
55
Question 56 0.5 pts
You conduct a study with four uncorrelated predictors. Then you remove one of the predictors. How should the removal of this predictor change the R2?
it will go up
it will go down
it will stay the same
there is not enough information to tell
Question 57 0.5 pts
The main difference between a simple and multiple regression is:
the number of dependent variables
the number of independent variables
the scales of measure of the independent variables
the size of the samples
Question 58 0.5 pts
In a curvilinear relationship:
the variables are unrelated
a scatterplot provides little information about the variables
the trend of the variables changes direction
an additional variable is necessary to explain the relationship
Question 59 0.5 pts
Another term for a regression intercept is:
unexplained variance
collinearity
heterogenity
constant
Question 60 0.5 pts
Nationwide, undergraduate college enrollment by gender is 55% female and 45% male. A professor wants to know the opinions of male versus female undergraduates nationwide. To ensure equal representation of the two genders, what sampling method should the professor use?
simple random
stratified random
cluster
convenience