Experimental Data

Question 1

What is the definition for generalizability?

Answer 1

The extent to which study results can be applied to other situations or people outside of the experiment.

Question 2

What is another term for generalizability?

Answer 2

External validity

Question 3

What is internal validity?

Answer 3

Has to do with how accurate and true the study is to the population it studies. Also paired with causality which asks if a change in the independent variable is really causing a change in the dependent variable?

Question 4

True or false:
Internal validity and external validity can be seen as working opposite each other.

Answer 4

True.
Internal validity is present when an experiment is tightly controlled and less generalized.
External validity is present when an experiment isn’t tightly controlled and more generalized.

Question 5

What are some threats to external validity?

Answer 5

An artificial research environment, non representative sample, small sample group, and measurement effects

Question 6

What are some threats to internal validity?

Answer 6

Confounding variables, selection bias, maturation, repeating testing, regression toward the mean

Question 7

What does it mean when a study is “reliable”?

Answer 7

It means the measure/study is consistent and produces similar results every time it’s repeated.
Other words for reliable are replicability/reproducibility.

Question 8

What does “correlation” describe?

Answer 8

It describes that there is or isn’t a relationship between variables.
Positive correlation means that when one variable goes up, the other also increases.
CORRELATION DOES NOT IMPLY CAUSATION. That is, the increasment in 1 variable directly causes the increasing of the 2nd variable.

Question 9

The correlation coefficient is termed “r”. If r is close to zero such as 0.05, describe the relationship between variables X and Y.

Answer 9

Variables X and Y have no correlation to each other, that is they have no relationship.
The closer “r” is to 0, the weaker the relationship.

Question 10

If r = + 1, relationship is (positive/negative).
If r = -1, relationship is (positive/negative).

Answer 10

1. Positive
2. Negative

Question 11

The correlation coefficient ranges from values of __ to __.

Answer 11

-1 to +1

Question 12

True or false:

Answer 12

A negative correlation coefficient means that variable aren’t related to each other at all.

Question 13

True or false:
Both observational and experimental studies can prove causation.

Answer 13

False, observational studies can’t prove causation, there are often times many confounding variables.

Question 14

What are confounding variables?

Answer 14

Uncontrolled variables that have an effect on the independent or dependent variable (results of experiment).

Question 15

Differentiate between two types of experimental trials: randomized control trial and nonrandomized design trial.

Answer 15

Randomized control trial - individuals are randomly assorted into treatment & placebo groups
Nonrandomized design - individuals are non randomly assorted into treatment & placebo groups

Question 16

Experimental trials are usually done for what purpose?

Answer 16

To determine efficacy of a certain treatment/intervention

Question 17

What is the difference between a longitudinal vs. cross-sectional study?

Answer 17

Longitudinal - data is gathered for a longer period of time, often to asses risk factors or outcomes/change in variables over time
Cross-sectional - an observational study where data is gathered only at one point in time for a population; used to measure prevalence of a disease/trait at a given point

Question 18

What are case studies (and their purpose)?

Answer 18

In-depth analysis and study which involve one or a few individuals who possess a trait/condition of interest

Question 19

What is a case-control observational study?

Answer 19

A study which includes data gathered from individuals with the condition of interest compared to a control group of individuals who don’t have the condition.

Question 20

What is sampling bias?

Answer 20

When participants are recruited from a population nonrandomly which can result in non representative samples that don’t reflect the overall characteristics of the population of interest.

Question 21

What are 3 different types of nonrandom study samples?

Answer 21

Convenience sample - a study uses the most easily accessible members of a population (for example, they use Chinese immigrant children to draw conclusions about Chinese children in general)
volunteer sample - a study uses only those who want to participate in the study
Snowball sample - a study uses people who have been referred to by participants already in the study

Question 22

What is the social desirability bias?

Answer 22

This is the tendency of participants and research studies to overemphasize positive behaviors, and underreport undesirable behaviors

Question 23

Which method is best to study complex human behavior: quantitative or qualitative method?

Answer 23

Qualitative method because it allows to subjectively gather information to determine the presence of a pattern

Question 24

What is an ethnography?

Answer 24

A qualitative method which allows researchers to study human behavior through observation & interviewing of people within their own communities in natural environments.
Ethnographies include info about culture, norms, values

Question 25

What is content analysis?

Answer 25

Used to examine words/text and images involved in communication such as looking at one-on-one verbal communication, interview transcripts, or online content/websites.
-Turns qualitative info into quantitative data for analysis

Question 26

What is social network analysis?

Answer 26

An epidemiological technique which maps connections between individuals (social ties/relationships) to study the spread of communicable diseases in a population.

Question 27

What does it mean to measure casual mechanisms?

Answer 27

It means to measure cause and effect relationships by manipulating independent variables to measure the effects on the dependent variables

Question 28

Temporality

Answer 28

Understanding causal relationships requires a sense of which events happen before others

Question 29

The tendency to agree with a statement when one is unsure is termed

Answer 29

acquiescence bias

Question 30

Which is usually higher, the incidence or prevalence rate?

Answer 30

Prevalence rate

Question 31

The tendency to answer questions in a way that makes us look better about ourselves

Answer 31

social desirability bias

Question 32

What is the difference between leniency bias and harshness bias?

Answer 32

Leniency bias refers to the tendency to answer every question leniently (that is, to be “too nice” in one’s answers).

Harshness bias refers to the opposite: the tendency to answer every question harshly, or be “too mean” in one’s answers.

Question 33

Define “demand characteristics” according to AAMC

Answer 33

These occur if the research design provides cues to the participants regarding the study hypothesis which influences their responses

Question 34

Name the three fundamental principles of research ethics

Answer 34

-Respect for persons
-Beneficence
-Justice

Question 35

Beneficence

Answer 35

The principle of maximizing the good that is done and minimizing the harm to participants.

Question 36

Justice

Answer 36

Justice refers to the obligation to distribute the burdens and benefits of research fairly.

Question 37

What is the Hawthorne effect?

Answer 37

When people change their behavior when they know they are being observed.

Question 38

What are opinion surveys?

Answer 38

Surveys which ask individuals predetermined questions to assess public attitudes about a topic.
“How do you feel about the economy?”

Question 39

Reading a box plot

Answer 39

Lower Quartile starts from the minimum value & ends at the edge of the box
Upper Quartile starts at the top edge of the box and ends at the max value
-Box plots provide information about median (NOT mean)
-If the vicinity of the median values (line in the middle of the box) of groups DON’T overlap, they are statistically significant

Question 40

Example question box plot

Answer 40

*Median values of group 1&2 overlap while group 3 doesn’t overlap with any

Question 41

How to tell if two groups are statistically significant from each other on a bar graph

Answer 41

Significant = error bars don’t overlap

Question 42

What is social epidemiology?

Answer 42

A subset of epidemiology which focuses on how social cultural factors contribute to disease patterns in populations.

Question 43

Takeaway concepts from question about concluding results from a study using experimental methodology
SB Psych #53

Answer 43

1. If the study includes a control group & experimental group & manipulates an independent variable to measure the dependent variable, results can show:
   -There is evidence that 1 factor causes the other (if experiment didn’t control for confounding factors)
   -There is a direct cause & effect relationship between the 2 factors (if experiment controlled for confounding factors)
2. If the study didn’t use experimental methodology, but used observational methodology [no actual manipulation of variables] results can show
   -A correlation relationship between the two factors

Question 44

What is counterbalancing and why do researchers use this in studies?

Answer 44

It is purposefully changing the order in which variables are presented, in order to control for the order (so this doesn’t affect the results).

Question 45

Define mean and median

Answer 45

Mean is the average value based on all the scores.
Median is splitting the sample of distribution values in half. If the median is 25, half of the values in the distribution are under 25 and half are over 25.