Task 6

Question 1

Sample

Answer 1

small subgroup/subpopulation chosen from the population.

Representativeness: a representative sample closely matches the characteristics of the population.

Biased sample: ratio of the two samples does not approximate that of the population.

Question 2

Generalization

Answer 2

ability to apply findings from a sample to a larger population. This increases the external validity of the study.

Question 3

Nonrandom sampling

Answer 3

used due to convenience, but they often differ from the general population, this may limit the ability to generalize and the external validity.
→ This is the standard procedure for animal research.

Question 4

Lab research

Answer 4

bringing participants into a lab environment that you created. Two ways of acquiring participants:
1. Ask for participants from whatever participant population is available.
2. Use a subject pool if this exists.

Participation must always be voluntary.

• When one is concerned with the loss of generality, they could use a simulation of a real-world situation. The researcher retains control over variables while studying the behavior under relatively realistic conditions. It is also less expensive and time-consuming than studying behavior in a field setting.

Question 5

Field research

Answer 5

conducting research in the participant’s natural environment.
Method of acquiring participants is dependent on the study.
Generally less control over who participates than in a lab.

Question 6

Volunteer bias

Answer 6

volunteers differ from nonvolunteers, this affects the external validity of research.
- participant related characteristics (education level, gender, etc.)
- situational factors (stress from study, incentives for participants, etc.)

Question 7

Volunteerism: internal validity

Answer 7

may affect inferred causality –> closely relates to internal validity
- desire to please experiment may cause attitude change

Question 8

Volunteerism: external validity

Answer 8

may affect the ability to generalize the results

Question 9

solutions for volunteerism

Answer 9

• make appeal interesting + non-threatening
• state the theoretical and practical importance
• monetary/gift rewards
• request made by person of higher status
• avoid stressful research

etc.

Question 10

Random sampling

Answer 10

every member of the population has an equal chance of appearing in the sample.

Question 11

Stratified sampling

Answer 11

provides a way to obtain a representative sample and make sure each segment of the population is represented.

1. Divide the population into segments or “strata”
2. Select a separate random sample of equal size from each stratum.

Proportionate sampling: variant of stratified sampling, proportions of people in the population are reflected in your sample.

Stratification and proportionate sampling can be done after collecting a sample.

Question 12

Systematic sampling

Answer 12

sampling every nth element from a population after a random start

Question 13

Cluster sampling

Answer 13

make sampling more manageable, identify naturally occurring groups and randomly select a certain cluster.

Saves time and is an acceptable, cost-effective method of acquiring a sample

Limits your sample to those participants found in the chosen clusters (possibility of leaving out important information)

Multistage sampling: variant of cluster sampling → identify large clusters and randomly select individual elements

This method can be combined with stratification procedures to ensure the representative sample

Question 14

Sample size

Answer 14

one should try to select an economic sample –> enough participants to ensure validity, but no more than that

consider:
1. amount of acceptable errors
2. expected magnitude of the population proportions

Question 15

Internal validity

Answer 15

ability of your research design to adequately test the hypothesis it was designed to test.

threatened by:
- rival hypothesis
- confounding
- history
- maturation: change in performance due to age/fatigue
- testing
- instrumentation
- statistical regression: subjects selected for treatment on the basis of their extreme scores tend to move closer to the mean on retesting (regression to the mean)
- biased selection
- experimental mortality

enhancing validity
- carefully plan
- identify plausible rival hypotheses
- redesign if necessary

Question 16

External validity

Answer 16

A study has external validity to the degree that its results can be extended (generalized) beyond the limited research setting and sample in which they were obtained

threatened by:
- reactive testing
- interactions between participant selection bias + independent variable
- reactive effects of experimental arrangements
- multiple treatment inferences.

highly controlled lab settings cause low external validity

Question 17

Regression towards the mean

Answer 17

tendency of extreme values in a dataset to move closer to the mean when measured again. This phenomenon occurs in situations where there is a random component or variability involved in the measurements.