Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

Introduction to research methods PSY10005 > Week 4 > Flashcards

Week 4 Flashcards

1
Q

Define “measurement”

A

Assigning a number to a property or construct.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What are the two properties of good measurement in psychometrics?

A

Reliability and validity.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What is reliability?

A

The consistently and stability of scores using your measurement instrument.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What is validity?

A

How well the measurement procedure is measuring what it is meant to be measuring AND how well those measurements are interpreted.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What is a reliability coefficient?

A

A score used to indicate the strength of reliability.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What is test-retest reliability?

A

The consistency of scores on a test over a period of time.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is equivalent-forms reliability?

A

The consistency of scores on two test measuring the same thing.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What is internal consistency reliability?

A

How consistently individuals items on a test measure a single construct.
The coefficient alpha/Cronbach’s alpha is the standard index, 0.70 or higher indicates that the items on the test are reliably measuring the same thing.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What is interrater reliability?

A

The consistency between two or more scorers.

Measured by interobserver agreement, which is the percentage of times different scorers agree.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What is validation?

A

Gathering evidence to support inferences made from test scores.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What is content validity?

A

Judgment by experts of the degree items or tasks on a test represent a construct. It is said to have content validity if it meets three criteria.

  1. It has face validity (the degree to which a test appears to measure what it is intended to measure)
  2. It does not underrepresent the construct (missing items/tasks).
  3. Do any items/tasks represent something other than the construct?
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Explain validity evidence, based on internal structure (for a measurement tool for a construct).

A

Determining validity by examining how well items in a test relate to the overall test score and if the test is measuring a unidimensional or multidimensional construct/trait.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What is factor analysis?

A

A statistical procedure that helps determine if items on a test are interrelated or are part of subsets, in summary it allows you identify if a test is unidimensional or multidimensional.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What is homogeneity?

A

The degree to which a set of items measure a single construct.
Two indices are used, item-to-total correlation and coefficient alpha.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What is validity coefficient?

A

A score used to indicate the strength of validity.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What is criterion-related validity?

A

The degree to which test scores predict performance or behaviour.
This includes:
1. Predictive validity - How well test scores predict future performance or behaviour in another related measure? (IQ -> ATAR)
2. Concurrent validity - How well test scores predict scores of another test measuring the same thing.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

What is convergent validity?

A

Validity evidence based on the degree to which the test scores correlate with other measures of the same construct.

18
Q

What is discriminant validity?

A

Validity evidence based on the degree to which the test scores do not correlate with the measures of a different construct.

19
Q

Explain known-groups validity evidence.

A

The variation of test scores in a group is proportionate to the known variation of the construct in the population.

20
Q

What is the norming group?

A

The reference group that the reliability and validity evidence is obtained from.

21
Q

What is a census?

A

Collecting data from everyone in the population (rare).

22
Q

What is a sampling frame?

A

The source material or list from which a sample is drawn.

23
Q

What is response rate?

A

The percentage of individuals who participant in the research study.

24
Q

What is proximal similarity?

A

That you can generalise research results to different populations to the degree that the sample is similar to the population.

25
Q

What are the four types of random sampling?

A
  1. Independent random sampling/equal probability of selection method (EPSEM). (either with replacement or without).
  2. Stratified random sampling.
  3. Cluster random sampling.
  4. Systematic sampling.
26
Q

What is stratified random sampling?

A
  1. Sample frames are divided into strata (categories).
  2. Draw random samples from each strata.
  3. Combine the samples from each strata to create the final sample.
27
Q

What is cluster random sampling?

A

A sampling method where clusters of individuals are selected as opposed to single individuals.

28
Q

What is systematic sampling?

A

A three step process.
1. You determine the sampling interval (k=N/n).
2. Randomly select a number from 1 to k and include that individual in the sample.
3. Select every kth individual for the sample.
(Can be problematic if there is a cyclical pattern in the sampling frame, this is called periodicity)

29
Q

What are main types of non-random sampling?

A
  1. Convenience sampling - Use of individuals who are readily available.
  2. Quota sampling - Setting quotas (number of kinds of individuals) then using convenience sampling to obtain a sample.
  3. Purposive sampling - Specify characteristics of interest, locate and select individuals who have the specified characteristics.
  4. Snowball sampling - Each sampled individual is asked to identify other potentional participant who meets the inclusion criteria.
30
Q

What is random assignment and why is it used?

A

Randomly assigning your sample into groups, it is used to produce two or more probabilistically equivalent groups.

31
Q

What are four rules to remember when determining sample size?

A
  1. If less than 100, use entire population.
  2. Larger sample sizes make it easier to detect an effect or relationship in the population.
  3. Compare with other studies by doing a literature review (what was the sample size?).
  4. Use a sample calculator (eg: G-power).
32
Q

What is research validity?

A

How correct an inference is, based off the research results.

33
Q

What is statistical conclusion validity?

A

The degree to which the IV and DV covary, the relationship is considered statistically significant if it is not likely due to chance.

34
Q

What is construct validity? (experimental research)

A

The degree to which a construct is accurately represented by it’s operational definitions.

35
Q

What are the threats to construct validity?

A
  1. Participant reactivity to the experimental situation - Participant perceptions and motives influence responses.
  2. Experimenter effect - Actions and characteristics of researchers that intentionally or unintentionally influence the responses of participants or the interpretation of results.
  3. Inadequate explanation, understanding or assessment of the construct - Do operationalisations represent constructs? Do they represent more than one construct? Are there enough operationalisations?
36
Q

What is internal validity?

A

The degree to which you can claim that the relationship between the IV and DV is causal.

  1. The IV must be related to the DV.
  2. Changes in IV must precede changes in DV.
  3. No other plausible explanation must exist for the effect.
37
Q

What are the threats to internal validity and some solutions to resolve with them?

A
  1. Extraneous variables - Confounding variables.
  2. History - Any event that influences the outcome other than the manipulation of the IV that occurs during the study before the post-test measurement.
  3. Maturation - Any physical or mental change that occurs between the manipulation of the IV and the observation of the DV. (Age, learning, fatigue, boredom, hunger etc).
  4. Instrumentation - Changes in pre-test and post-test assessment or measurement.
  5. Regression artifact - Effects that are due to regression to the mean. (Extremes scores regress to the mean).
  6. Attrition - Participant drop out.
  7. Selection - Biased selection or lack of random selection.
    * All these effects can interact*
38
Q

What is external validity?

A

The degree to which the study results can be generalized to other people, settings, treatments, outcomes, and times.

39
Q

What are the four major types of validity? (in experimental research?)

A
  1. Statistical.
  2. Construct.
  3. Internal.
  4. External.
40
Q

If you can’t eliminate confounding variables how can you control them?

A

Constancy; ensuring the confounding variable has the same effect on all groups (equating the groups).

41
Q

What are the five types of external validity?

A
  1. Population validity - Degree to which results can be generalised to population (accessible or target).
  2. Ecological validity - Degree to which results can be generalised across environmental conditions.
  3. Temporal validity - Degree to which results can be generalised across time.
  4. Treatment variation validity - Degree to which results can be generalised across variations in the manipulation of the IV.
  5. Outcome validity - Degree to which results can be generalised across related DV’s?
42
Q

What are the two kinds of cluster sampling?

A
  1. One-stage cluster sampling, in which the final sample is all of the clusters selected.
  2. Two-stage cluster sampling, in which the final sample is randomly selected from the clusters selected.

Introduction to research methods PSY10005 (10 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions