Theory: evidence suggests it to be true, but not proven

Hypothesis: educated guess

Reliability-necessary but insufficient characteristic of validity; consistency of measurement (scales)

Validity-the extent to which our inferences/conclusions are meaningful, useful, and appropriate. Validity is not a characteristic of a test itself; it is characteristic of the use of a test ex. Eye chart is valid use for whether or not someone can see, not how well they do in college

Nominal: named categories (race, gender, religion)

Ordinal: ordered categories (grade level, 5-star rating)

Interval: equal intervals on scale (temperature in Fahrenheit and Celsius)

Ratio: equal intervals and absolute zero (height, weight, time, scores, income)

z-score=x-x(bar)/SD group

effect sizes=mean of experiment-mean of control/SD control

How closely related two things are, closer to one is closer correlation; can be very effected by outliers;

Draw scatterplot, think about outliers, correlation coefficient, outliers make cloud bigger

Autonomy-respect for persons, sign informed consent, freedom of choice to be in the study or not, protection

Justice-ensure that participants understand value and receive benefits, have lots of different people involved so we can extent the results to many; don’t do AIDS study in Africa because we have a lot of people with AIDS and then treatment is too expensive for them

Beneficence- we protect participants by minimizing risks, weighing risks and benefits (physical or psychological, think Stanford prison study/milgram experimentàis it worth it?

Ethnography-study of culture, defined by groups of people that share beliefs and behaviors; ex. College students, vandy, HOD, greek, race, religion

Case study-bounded study; beginning and end, study a specific time or event, geographical limits; ex. Woodstock, rush 2011, professor, organization, HOD dept.

Phenomenological-studying a certain phenomenon; ex. Prejudice because it knows no boundaries (no geographic or time boundaries); ex. Indiscretions by married men and then Tiger Woods case study

Grounded theory-theories emerge from qualitative research; example the 4 frames by bowman and deal

Rating scale: 1-5

Rank: favorite to least favorite

Semantic differential: bipolar words (good/bad; pretty/ugly)

Likert scale: strongly agree, agree, neither agree or disagree, disagree, strongly disagree (no basis for judgment)

Multiple choice: a,b,c,d

“poll” that leads you to believe they are interested in your opinion, really skewed information that is a psychological strategy to persuade (don’t ask about demographics)—did you know cigarettes kill millions of people per year? Do you support legislation?

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

1. What’s the risk in believing something simply because it’s obvious?

1. What’s the difference between correlation and causality? What’s the benefit of correlational data?

1. What is required to present a strong case for a causal relationship between two variables?

1. What is required before you can say you truly know something? Provide some examples.

1.     has to be true; 2. Have to believe it is true 3. Have to be justified in knowing

weather, 

1. Know the five (5) key rules for creating professional graphs (and be able to fix a flawed graph).

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

Quantitative: variables, relationships between variables, and who population to be studied is

Qualitative: central phenomenon, participants, setting

1. 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?

Sampling error: inevitable, never exactly correct measurement of a population unless you take a census; helps to increase sample size

Sampling fluctuation: change in values when sample population changes due to sampling error (normal fluctuation)

Statistical significance: when a value is more than different than due to typical sampling fluctuation

1. What are statistical significance, effect size, and practical significance? What do they tell us?

Statistical significance: mathematical test that gives yes/no answer to the question: are the differences we observe larger than we would expect from sampling fluctuation alone?

Effect size: magnitude and direction of the differences (mean of experiment-mean of control group/SD control)

Practical significance: answers so what?, tells us how important differences are in terms of what people value

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

Quantitative: goal is to get a representative sample of the population you want to generalize your results to; strategies, random, systematic (every nth person), proportional stratified (different races but proportional to actual proportion); disproportional (10 of each race)

Qualitative: goal is to get the most informed participants; strategies are maximum variation (both ends of the spectrum), snowball sampling, typical case (QVC stereotype-typical food network viewer)

1. 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.

1. What are five (5) characteristics of good questionnaire questions?

-people understand your question (do a pilot study to check)

-standardized (same amount of time for survey, how you hand it out is the same)

-indicates the type of response required (yes/no or longer answer)

-willing to answer (not about sex/drugs)

-able to answer (have the info)

1. What are five (5) reasons for conducting a review of the literature?

-gives you information to make an educated guess (hypothesis)

-gives credibility to your study

-identify contradictory findings

-learn from the limitations of other studies/use same instruments

-find things you never expected to know

1. What are five (5) strategies you can use to increase the reliability of your observations?

1.     multiple observers

2.     training

3.     strive for inter-rater reliability

4.     use low inference measures (objective)

5.     triangulation (multiple measures of the same thing)

1. What are five (5) strategies you can use to increase the credibility of your qualitative research?

1.     use triangulation-multiple methods to collect data

2.     prolonged and prolific fieldwork

3.     gather a lot of fieldnotes (rather than memorizing)

4.     verbatim quotes

5.     number checking

6.     be a participant observer

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

1.     what you’re measuring (physical, cognitive, affective)

2.     technique (reading graduated cylender)

3.     instruments (scales at home vs. doctors) 

1. Why would you choose an ex post facto research design to answer a question? How would you go about conducting it? (Make sure to describe in detail the most critical design element.)

-when it is unethical, illegal, immoral, impractical to do traditional research

-establish cause and effect relationship without conducting actual experiment; control extraneous variables and then manipulate chosen variables

matching (women, same SMS, highly educated, age 30)

1. How are the results of a typical single subject research graphed? Provide an example and use it to describe the major elements of this research design.

ABA(B) format—baseline, treatment, remove treatment

Low, high, decreasing back to baseline (sometimes give treatment again to leave subject with good treatment)

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

1. List the 11 threats to internal validity. Name one, 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.

1.     history

2.     experimental mortality

3.     statistical regression to the mean

4.     maturation

5.     instrumentation

6.     testing

7.     selection

8.     diffusion

9.     compensatory rivalry

10.  compensatory equilization

11.  demoralization

testing-become test wise rather than ability improving (like an SAT prep class); use only a few tests, non-reactive tests

1. What would it take to change your mind about something you have a strongly-held belief for?

Need to be presented with new information that is stronger than what you’ve previously been given

Reasons for knowing: 1. Be truth 2. Believe it is truthful 3. Reliable source

1. Descriptive, comparative, correlational, quasi-experimental, and true experimental research

Descriptive: 1 person

Comparative: 2 groups and 1 variable, ex. Men’s vs. women’s love of chocolate

Correlational: 1 group with 2 variables, height and weight (everything is correlated somehow)

Quasi-experimental: non-random assignment of participants to groups, variable manipulated

True experimental: typical research, random group selection, introduction of a variable