Midterm Exam

Question 1

Theory

Answer 1

a model to explain what we observe. All theories are wrong

Question 2

Hypothesis

Answer 2

a clear, concise, testable statement. Similar to educated guess.

Question 3

Descriptive Research

Answer 3

information about the frequency or amount of something (mean, median, mode) (ex: statistics on new Vanderbilt class)

Question 4

Correlational Research

Answer 4

descriptions of the differences between groups; compare two groups on one variable (ex: difference in preference between men and women between chocolate preferences)

Question 5

Comparative Research

Answer 5

description of the relationship between variables (ex: the relationship between height and weight)

Question 6

Experimental Research

Answer 6

an investigation characterized by the direct manipulation of one variable (the cause) so its effect can be seen on another variable (the effect) while controlling for other extraneous variables

Question 7

Be able to write an APA 6th edition citation for a peer-reviewed journal publication.

Answer 7

Author, A.A., & Author, B.B. (year). Title of article. Title of Journal, volume #, nn-nn.

Question 8

Internal validity

Answer 8

the extent to which the independent variable, and not other extraneous variables, produce the observed change in the dependent variable (ex: checking weight on scale)

Question 9

External validity

Answer 9

the extent to which the results of a study can be generalized to other subjects, setting and time

Question 10

z-score

Answer 10

the relationship of one score to the norm group in terms of standard units; how many standard deviations above or below mean is a score; z=(raw score-mean)/SD

Question 11

Effect size

Answer 11

a measure of the magnitude and difference of the means of two groups. ES=[Mean(control) - mean (experimental)]/SD(control). Effect size helps readers understand the significance of the results whereas p-values only tell if an effect exists.

Question 12

Correlation Coefficient

Answer 12

Correlation - a measure of the strength of the relationship between two variables. Strength 0 to 1, 0 indicates no relationship (no predictive power), 1 indicates a perfect relationship (perfect predictive power); direction: positive or negative. Straight line = 1.00, Corn on the Cob = 0.75, Football = 0.5, Circular = 0.0 An outlier will move the correlation coefficient closer to zero.

Question 13

Words of estimative probability

Answer 13

words that convey your degree of confidence or certitude in very clear terms. (Impossible, Unlikely, 50/50, Likely, Certain). Important for analysts or people making decisions to tell someone and convey to them clearly your understanding of the fact

Question 14

Weasel Words

Answer 14

words that don’t convey at all with any degree of certitude or confidence what the likelihood of something is.

Question 15

What are the ways of knowing we discussed in class? Why are some ways better than others?

Answer 15

1. Personal Experience - most people weight the most, but only 1 data point out of thousands and could be outlier; people can have very different perceptions (eyewitness testimony is notoriously poor, so can’t be very accurate)
2. Reason and Intuition - logic and formal reasoning, but doesn’t make it true
3. Authority - just because someone has Ph.D. doesn’t mean its true or not believing because of who someone is
4. Tradition - finding answers through long-established customs, but based on idealized past so possibly not best way moving forward
5. Systematic Inquiry - systematic process for collecting and processing information. Systematic, testable and objective and allows us to examine process as well as results. Not all research is equally good so helps discern good research.

Question 16

What’s the risk of believing something simply because it’s obvious?

Answer 16

Might be fooled. People have a tendency to believe any reasonable statement about human behavior. Baratz’s study shows this. If someone provides a reasonable statement with some explanation that sounds logical, people will buy into it. Then if have data with correlation coefficient, then they will really believe it.

Question 17

What’s the difference between correlation and causality? What’s the benefit of correlational data? What’s required to make causal claims? What is the risk of confusing these two?

Answer 17

Correlation is the relationship between two variable. Benefit is you can use one variable to predict another. Causality is best established through experimental research design. Correlation doesn’t give evidence for cause and effect relationship. People get tricked because they can think of causes for relationship. One doesn’t tell us anything about the other. Could be a cause and effect relationship, but if low correlation then can’t be the cause.

Question 18

List the required elements of qualitative and quantitative research questions and give examples

Answer 18

• Quantitative Research: Identifies specifically: 1) the variables, 2) the relationship between them, and 3) the subjects. Can be descriptive, correlational/comparative, or experimental (causal). (what is the relationship between…)
• Qualitative Research: identifies specifically: 1) the central phenomenon, 2) the participants, and 3) the site. (perceptions of…)

Question 19

How are sampling error, sampling fluctuation, and statistical significance related? How can you reduce sampling error and increase the likelihood your results will be statistically significant?

Answer 19

Sampling Error: the value for the sample is not equal to the true population value. Reduce by increasing size of sample
Sampling Fluctuation: when take two samples from same population. When difference in samples, larger than expect from sampling fluctuation, say it’s a statistically significant difference
Statistical Significance: a mathematical test that gives a yes/no answer to the question” “Are the differences we see larger than we would expect than from sampling fluctuation alone?”. Increase likelihood of results being statistically significant by increasing number of people in your study

Question 20

How are theories and hypotheses related?

Answer 20

hypotheses are generated from theories and to the extent that we provide evidence from the hypothesis that’s true is true our theories support it

Question 21

What are statistical significance, effect size, and practical significance? What does each tell us?

Answer 21

Statistical Significance: a mathematical test that gives a yes/no answer to the question: are the differences in outcomes we observe larger than from sampling fluctuation alone?
Effect Size: tells us the magnitude and direction of the difference between two groups/samples of data. Tells us if one is higher than the other and by how many standard deviationsPractical Significance: answers the question: so what? What does it mean?
Ex: Vanderbilt football

Question 22

What are the goals and strategies (3 each) for sampling in qualitative and quantitative research?

Answer 22

Qualitative: get the most informed participants
1. use network/snowball sampling
2. typical case (representative participant)
3. extreme case (unique/atypical participant)
Quantitative: to get a representative sample of the population you want to generalize the results to
1. simple random sampling
2. stratified sampling (proportion of subjects in each strata in the population are reflected in the proportions of subjects in each sample strata)
3. cluster sampling (district, school)

Question 23

What’s so good about pretest-posttest control group experiments? Be able to draw a diagram with the key elements and label the purpose(s) of each in the context of a real-world study.

Answer 23

Pretest-Posttest Control Group Experiments

• Control Group: To help reduce threats to internal validity. This is not required of experiments but is very important.
• Randomly Assigned Subject: to help ensure equivalence between the two groups…on the dependent measures as well as all others
• Pretest: to test for equivalence of the groups at the start and for baseline data to calculate pretest/posttest delta
• Subject Treatments: the experimental group gets the treatments, and the control group gets something unrelated to the DV
• Posttest: measure the delta between pretest and posttest and measure the delta between groups on the posttest
• Delayed Retention Test: to determine whether the effects are lasting or whether they fade quickly
• Experiment Specific information: subjects, dependent variable, treatments

Question 24

What are the 7 Survival Skills?

Answer 24

1. Critical Thinking and Problem Solving
2. Collaboration Across Networks and Leading by Influence
3. Agility and Adaptability
4. Initiative and Entrepreneurialism
5. Effective Oral and Written Communication
6. Accessing and Analyzing Information
7. Curiosity and Imagination

Question 25

Stephan J. Gould Quote

Answer 25

- Correlations are really great things, allow us to predict one variable given another which is very useful - Causality is where there is a cause and effect relationship between two variables - one actually causes the other, and the best strategy for showing this is a pretest-posttest - Unfortunately, many people believe that evidence of correlation is evidence of causality when there actually isn't any evidence for that - Cleaver people are good at coming up with plausible explanations - So what? Then politicians, legislators or adults try to create rules or laws that prohibit people from doing things to get cause-and-effect benefit when only correlational evidence exists - So on many issues, people use correlational data to try to provide an explanation for data that's correlational and enact legislation and that’s a mistake

Question 26

Be able to correctly draw a graph provided to you that is inaccurate.

Answer 26

1. Choose the right graph for your data (bar>pie) 2. Title or caption, label axis, and include units (use horizontal text wherever possible, title is largest, then labels, then units) 3. No color, shading/gradients, or 3D (if use color, only use one) 4. Be consistent from graph to graph 5. The less ink, the better - keep it simple 6. Start number at origin

Question 27

What three (3) factors influence the reliability of your measurements? (Provide examples.)

Answer 27

1. What you measure (physical traits, cognitive traits, affective traits) (height, weight, blood pressure) 2. the instrument you use (measurement of BMI: height/weight table, bioelectrical impedance, skinfold, dual energy x-ray absorptiometry) 3. your technique (meniscus reading)

Question 28

What counter-argument are you trying to minimize by controlling for threats to internal validity?

Answer 28

"Isn't it possible that the difference in outcomes you saw between the control group and the experimental group was not a result of the treatment, but rather it was the result of ____?" (Fill in blank with 11 threats to internal validity)

Question 29

List the 11 threats to internal validity.

Answer 29

"HERMITS DRED" 1. History (coincidental events) 2. Experimental Mortality (Attrition) 3. Statistical Regression to the Mean 4. Maturation 5. Instrumentation 6. Testing 7. Selection ("Assignment") 8. Diffusion 9. Compensatory Rivalry 10. Compensatory Equalization 11. Demoralization

Question 30

Name one threat to internal validity, define it, provide a real world example of how it could affect a study, and describe three (3) strategies you could use to minimize it as a threat.

Answer 30

Maturation: subjects naturally mature (physically, cognitively, or emotionally) during the course of an experiment - especially long experiments. Example: If do a program to help young children learn to run faster over a few weeks, could be because they grew and got stronger because kids develop physically fast as opposed to being a result of the program. 3 Strategies you can use to minimize threat: 1) use a control group, 2) Keep the study as short as possible, and 3) investigate beforehand the anticipated effects of maturation.