Show:
Chapter 15!
More Chapter 15!
A Sample Scenario!
Another Scenario!
Chapter 5 Study Guide
100
Which of the following is the correct definition of "power?"
A.) The probability of rejecting the null hypothesis when it is false
B.) The probability of rejecting the null hypothesis when it is true
C.) The probability of rejecting the null hypothesis when you have at least 30 subjects in the study
D.) The probability of rejecting the null hypothesis when you have less then 30 subjects in the study
Power is . . .

A.) The probability of rejecting the null hypothesis when it is false
100
When the dependent variable in an experiment can be measured using a nominal scale of measurement, which statistical test should be used?

A.) Mann-Whitney U
B.) Chi-Square
C.) Independent-Samples T-Test
D.) Dependent-Samples T-Test
B.) Chi-Square
100
Dr. Fore-Gettes studies the factors that affect test taking. He believes people perform better on tests when the temperature of the room is cold. He also believes that people perform better on tests when they take it in the same room in which they learned the material. To test his ideas, he has participants study a list of 20 facts in a room of his laboratory. After a short break, during which participants leave the room, he gives them a test of recall. The test takes place in either the same room or a different room from where the participants studied. For half of the subjects in each room, the thermostat is set to warm and for the other half it is set to cold.

If Dr. Fore-Gettes found a significant main effect of testing room, and a significant main effect of room temperature, but NO INTERACTION, what is one possible value that could be present in the table for this experiment? (I will draw the table for the experiment on the board).
80
100
A social psychologist wanted to test the hypothesis that the eating behavior of overweight people is more influenced by external factors than internal factors (e.g., actual physiological stimuli associated with hunger), as compared to the eating behavior of non-overweight people. As a way of addressing this possibility, college students who had signed up to be in a memory experiment were first brought into a waiting room where they were asked to remain until called for the experiment. As they entered the waiting room, raters observing behind a one-way window classified subjects as being members of one of two groups based on their physical appearance: “overweight” or “non-overweight.” On the wall of the waiting room was a clock that, for one group of subjects, was set to run fast so it appeared to be dinnertime not long after the subjects had arrived in the room. For another group of subjects, the clock was set to run slow. For this group, dinnertime was actually approaching, but it did not appear to be by the time indicated on the clock. Also, in the room, were bowls of crackers and subjects were invited to help themselves to crackers if they wished. The researchers’ hypothesized that the eating behavior of the “non-overweight” subjects would not be influenced by the time indicated on the clock whereas the eating behavior of the “overweight” subjects would be.

Identify whether the study described is corrleation, experimental, or some type of quasi-experimental design.
Quasi-Experimental Design
100
What is a confounding variable? Why is this important?
A confounding variable is one whose effect cannot be separated from the supposed independent variable. This is important because in science, we want to rule out alternative explanations, and confounding variables prevent us from drawing decisive conclusions from our study’s results.
200
Power is equal to . . .

A.) The difference between the means of the groups, divided by the pooled estimate of the population standard deviation
B.) The standard deviation of the population divided by the square root of N
C.) 1 - the probability of a type II error
D.) 1 - the probability of a type I error
Power is equal to . . .

1 - the probability of a type II error
200
What is the textbook's definition of "statistical significance?"
This can be found on p.394!
"The probability that an experiment result happened by chance."
200
For this study, the statement, "Students perform better in a cold room than in a warm room" describes which of the following?

A.) A main effect of testing room
B.) A main effect of room temperature
C.) A simple effect of testing room location
D.) A simple effect of room temperature
E.) Both B & D
B.) A main effect of room temperature
200
A social psychologist wanted to test the hypothesis that the eating behavior of overweight people is more influenced by external factors than internal factors (e.g., actual physiological stimuli associated with hunger), as compared to the eating behavior of non-overweight people. As a way of addressing this possibility, college students who had signed up to be in a memory experiment were first brought into a waiting room where they were asked to remain until called for the experiment. As they entered the waiting room, raters observing behind a one-way window assigned subjects to one of two groups based on their physical appearance: “overweight” or “non-overweight.” On the wall of the waiting room was a clock that, for one group of subjects, was set to run fast so it appeared to be dinnertime not long after the subjects had arrived in the room. For another group of subjects, the clock was set to run slow. For this group, dinnertime was actually approaching, but it did not appear to be by the time indicated on the clock. Also, in the room, were bowls of crackers and subjects were invited to help themselves to crackers if they wished. The researchers’ hypothesized that the eating behavior of the “non-overweight” subjects would not be influenced by the time indicated on the clock whereas the eating behavior of the “overweight” subjects would be.

If you decide that the design is correlational, identify the predictor and criterion variables and indicate the type of scale on which they are measured.

If you decide that the design is experimental or a type of quasi-experimental design, indicate on what basis you made this decision and identify the relevant variables.
Speed o the clock is manipulated by the researcher; thus, it's a true IV. Subjects, however, cannot be randomly assigned to different levels of the weight variable, as they are simply assigned based on their appearance. So we have one between-subjects variable and one "subject variable" which is not manipulated.
200
What is the difference between a continuous variable and a discrete variable?
Continuous variable: one that falls along a continuum and is not limited to a certain number of values. Discrete variable: one that falls into separate bins with no intermediate values possible.
300
What is "Type I Error?"
Answer it in your own words, and then draw a diagram on the board for the class.
The probability of rejecting the null hypothesis when it is actually true
300
Which of the following is NOT NECESSARILY TRUE when you find a statistically significant result?
(i.e., what should you NOT NECESSARILY CONCLUDE if you have a statistically significant result?)

A.) That the effect of the independent variable was large
B.) That the results are quite important
C.) That the effect was large enough that the probability that it happened due to chance was quite small (i.e., less than 5%)
D.) A and B are not necessarily true
E.) All of the above are not necessarily true
A & B!
300
The following statistical test(s) should be administered to determine whether Dr. Fore-Gettes's findings are statistically significant:

A.) An independent-samples (between-subjects) t-test
B.) A dependent-samples (within-subjects) t-test
C.) A one-way ANOVA followed by appropriate posttest comparisons
D.) A two-way ANOVA followed by appropriate posttest comparisons
E.) A three-way ANOVA followed by appropriate post-test comparisons
D.) A two-way ANOVA followed by appropriate posttest comparisons
300
A social psychologist wanted to test the hypothesis that the eating behavior of overweight people is more influenced by external factors than internal factors (e.g., actual physiological stimuli associated with hunger), as compared to the eating behavior of non-overweight people. As a way of addressing this possibility, college students who had signed up to be in a memory experiment were first brought into a waiting room where they were asked to remain until called for the experiment. As they entered the waiting room, raters observing behind a one-way window assigned subjects to one of two groups based on their physical appearance: “overweight” or “non-overweight.” On the wall of the waiting room was a clock that, for one group of subjects, was set to run fast so it appeared to be dinnertime not long after the subjects had arrived in the room. For another group of subjects, the clock was set to run slow. For this group, dinnertime was actually approaching, but it did not appear to be by the time indicated on the clock. Also, in the room, were bowls of crackers and subjects were invited to help themselves to crackers if they wished. The researchers’ hypothesized that the eating behavior of the “non-overweight” subjects would not be influenced by the time indicated on the clock whereas the eating behavior of the “overweight” subjects would be.

If a dependent variable was used, identify it and the scale on which the values produced by the measure of the DV would lie.
The dependent variable is the number of crackers consumed, and this measure would produce values on a ratio scale
300
What is “reliability?” What is “validity?”
Reliability: The property of consistency of a measurement that gives the same result on difference occasions Validity (of a measurement): the property of a measurement that tests what it is supposed to test
400
What is "Type II Error?"

Answer in your own words, and then draw a diagram on the board for the class.
The probability of deciding that the null hypothesis is true when it is actually false
400
What are some reasonable working definitions of the terms "null hypothesis" and "alternative hypothesis?"

Also, create a hypothetical example scenario that accurately sets up a null hypothesis and an alternative hypothesis in an experiment.
Null hypothesis: The hypothesis that is of no scientific interest; sometimes called the "hypothesis of no difference." I sometimes think of this as, "nothing's really going on that's interesting."

Alternative hypothesis: The research hypothesis. The hypothesis which posits a difference.
400
Let's say that in one version of this experiment, Dr. Gettes has all subjects go through both of the levels of the Testing Room Variable (Same/Different), but he has different groups of subjects go through both of the levels of the Room Temperature Variable (Warm/Cold).
Which term would appropriately describe this type of experimental design?
A.) A within-subjects factorial design
B.) A between-subjects factorial design
C.) A mixed-subjects factorial design
D.) None of the above
C.) A mixed-subjects factorial design
400
A social psychologist wanted to test the hypothesis that the eating behavior of overweight people is more influenced by external factors than internal factors (e.g., actual physiological stimuli associated with hunger), as compared to the eating behavior of non-overweight people. As a way of addressing this possibility, college students who had signed up to be in a memory experiment were first brought into a waiting room where they were asked to remain until called for the experiment. As they entered the waiting room, raters observing behind a one-way window assigned subjects to one of two groups based on their physical appearance: “overweight” or “non-overweight.” On the wall of the waiting room was a clock that, for one group of subjects, was set to run fast so it appeared to be dinnertime not long after the subjects had arrived in the room. For another group of subjects, the clock was set to run slow. For this group, dinnertime was actually approaching, but it did not appear to be by the time indicated on the clock. Also, in the room, were bowls of crackers and subjects were invited to help themselves to crackers if they wished. The researchers’ hypothesized that the eating behavior of the “non-overweight” subjects would not be influenced by the time indicated on the clock whereas the eating behavior of the “overweight” subjects would be.

Identify how the data collected in this study should be analyzed.
Data should be analyzed with a two-way between-subjects ANOVA followed by appropriate posttests if a significant interaction of the two variables is obtained.
400
What is construct validity?
A test where the measurements actually measure the constructs they are designed to measure, but not others.
It concerns the question of whether the results support the theory behind the research. In other words, can you generalize from the specific operations of your experiment (including people and settings) to the general theoretical construct about the population in question?
500
What three things influence the power of a test, and how do each of these three factors influence power?
You must get all three things to get the points for this question!
As you announce these things, I will create a diagram on the board to demonstrate the concepts
1.) The value of alpha (the smaller your alpha, the smaller your power) 2.) The size of your experimental effect (as the experimental effect gets larger, the power increases) 3.) The size of your sample (increasing the size of the two groups used decreases variability of the sampling distribution , which reduces the amount of overlap between them)
500

Two part question:

1.) If I have an experiment with one independent variable and three or more levels for this independent variables, AND each participant experiences every condition, which statistical test should I use to analyze the data?

2.) If I have an experiment with two independent variables, each with two levels, and each subject only participates in one condition in the experiment, which statistical test shoudl I use to analyze the data?

1.) A one-way within-subjects ANOVA

2.) A two-way between-subjects ANOVA

500
I couldn't think of a fourth question on the scenario in this column, so instead I'm going to quiz you on directional hypotheses! For the two following statements, identify the correct null hypothesis and the correct alternative hypothesis. Have someone come to the front of the class to write these on the board for all to see.

1.) Introductory algebra students who are taught with hand-held calculators for a 15-week period will have different scores on tests of computational skills (taken without use of a calculator) from introductory algebra students who are taught without calculators (group 2).

2.) New graduate male nurses whose orientation program uses a preceptor (an individual tutor) as a major component (Group 1) will exhibit higher performance levels than will new graduate male nurses in a traditional orientation program (Group 2.)
The first should show a null hypothesis of the two groups being equal, and an alternative hypothesis that they are not equal.

The second should show a null hypothesis that the male nurses with a preceptor exhibit scores lower than or equal to Group 2, while the alternative hypothesis is that they will score higher than Group 2.
500
A social psychologist wanted to test the hypothesis that the eating behavior of overweight people is more influenced by external factors than internal factors (e.g., actual physiological stimuli associated with hunger), as compared to the eating behavior of non-overweight people. As a way of addressing this possibility, college students who had signed up to be in a memory experiment were first brought into a waiting room where they were asked to remain until called for the experiment. As they entered the waiting room, raters observing behind a one-way window assigned subjects to one of two groups based on their physical appearance: “overweight” or “non-overweight.” On the wall of the waiting room was a clock that, for one group of subjects, was set to run fast so it appeared to be dinnertime not long after the subjects had arrived in the room. For another group of subjects, the clock was set to run slow. For this group, dinnertime was actually approaching, but it did not appear to be by the time indicated on the clock. Also, in the room, were bowls of crackers and subjects were invited to help themselves to crackers if they wished. The researchers’ hypothesized that the eating behavior of the “non-overweight” subjects would not be influenced by the time indicated on the clock whereas the eating behavior of the “overweight” subjects would be.

Indicate any possible problems with this study!
1.) A measure of inter-rater reliability for weight classifications should be obtained 2.) Also, how was the DV recorded? The method should be explained in greater detail. 3.) Anything else????
500
What are the three elements of construct validity? (this is related to the 400pt question! :) )
(1) It should measure whatever theoretical construct it supposedly tests
(2) It should measure what it intends to measure but not measure theoretically unrelated constructs
(3) It should prove useful in predicting results related to the theoretical concept it is measuring.