Lecture 2 Flashcards

1
Q

What are the 6 steps to test construction?

A
  1. Goal
  2. Definition (of construct and how it will be measured)
  3. Construction (or using pre-existing test after doing research literature)
  4. Reliability
  5. Validity
  6. Utility (was it useful? did it provide the wanted information?)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What are the objectives and procedures for the goal of the measure?

A

-objective: what do we want to measure?; Among whom? (target population); in which context?; to what end? To solve which problem?
-procedures: personal values and experiences; readings; expert advice

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

What are the objectives and procedures for the theoretical definition?

A

-objectives: defining the construct; identify/defined its main components; clear indicators
-procedures: readings, lit review; brainstorming; consultation of theoretical and practical experts

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

What are the 3 levels of definition?

A
  1. global construct: global but precise definition of construct
  2. specific components: defining key components of construct
  3. behavioural indicators: definition of behavioural indicators of each dimension
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What are behavioural indicators?

A

-some indicators can be worth more than others
-consider consistent and inconsistent (opposite, not just adding “not”) indicators to cover the whole range of the construct
-differentiate core indicators from peripheral indicators

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

What are the objectives and procedures of construction?

A

-objectives: select the most appropriate type of measurement instrument
d scale; write items; try to prevent sources of error
-procedures: standardization of test administration process; writing clear instructions; focus group (test)

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

What are the 2 big types of tests?

A

-maximal performance/ability tests (knowledge, skill achievement, aptitudes, etc.)
-typical performance (attitudes, personality, behaviours, feelings, etc.)

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

What is a maximal performance test?

A

-right or wrong answers
-maximal level of performance (everything correct)

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

What is a typical performance test?

A

-no right or wrong answers
-daily life characteristics
-personality, interests, attitudes, etc.
-aim to describe the respondent as they are, with no assumption that any score is “better” than any other score

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

What are the 2 classifications of maximal performance tests?

A
  1. achievement (current acquisitions, current level of mastery/knowledge) OR aptitude (future mastery potential)
  2. power (items of varying levels of difficulty) OR speed (easy but numerous items)
    -ex: typing test (speed + achievement); school exam (power + achievement); intelligence test (power + aptitude); manual dexterity (speed + aptitude)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What are the 2 types of typical performance tests?

A

-objective: structured questions and responses format
-projective: unstructured/ambiguous questions, based on the assumptions that respondents will “project” their natural spontaneous characteristics in their responses

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What are the 3 types of administration modes?

A

-individual: assessor works with 1 person at a time; more flexibility in terms of content and responses; motivation; spontaneity & accuracy; time & cost
-group: rigidity(needs to be same administration); time & cost (less expensive and long)
-computerized (falls in between: no physical assessor, flexible, etc.)

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

What are the 2 types of expected results for all tests?

A

-norm-referenced: in comparison with the levels typically observed among members of a specific population (average, or below/above)
-criterion-referenced: in comparison with some pre-established level of performance (ex: DSM diagnosis)

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

What are the 2 types of expected results for tests measuring multiple dimensions (norm or criterion-referenced)?

A

-normative: absolute strength of the various characteristics (comparison of the strength of each specific characteristic across individuals)
-ipsative: relative strength of the various characteristics (comparison of the relative strength of various characteristics for a single individual) - formed choices

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

What are the response formats of maximal performance tests?

A

-true/false
-matching
-multiple choices
-open-ended questions (sentence completion; short response; essay)

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

What are the differences between response choices with open-ended questions?

A

-similarities: assessment understanding and application; [kinda for mc] assessment analysis, synthesis, reasoning; [kinda for essay] items response speed
-differences (essays): assessment originality; item writing facility (easier for creator); item free from guessing
-differences (multiple choices): inter-rater reliability; computerized scoring

17
Q

What are some additional considerations …

A
  1. difficulty of the items (should the test have varying difficulty; difficult items?)
  2. favour or avoid certain types of items (consider previously successful items; consider population)
  3. speed limit (should there be one; severe or not; is it an objective of the measure; is it affected by population)
18
Q

What are the response formats for typical performance tests?

A

-true/false
-forced choices (ipsative tests)
-likert scales (4-5 points [max 7], report level of agreement in interval measure interval-level measurement)
-continuous scales (2 extremes, ex: 1 to 10)

19
Q

What are the response formats for typical performance tests (2)?

A

-Q-sort (ex: sort cards into 9 sets, from least relevant to most relevant) [not very used]
-Scales with behavioural anchors (as clear as possible with each response scale; a lot longer to complete)
-Bipolar scales / semantic differentiator

20
Q

What are the response formats for typical performance tests (3)?

A

-projective: pictorial (ink blots, TAT); verbal (free association); drawing (draw a person, family, house [for kids])
-ranking scales (ranking many individuals): full; forced (4 best; extreme); paired (rank pairs)
-sociometrics (studies focusing on social network; ex: asking students to rank each other [popularity])

21
Q

What are the item writing rules and principles?

A
  1. items should be easy to understand by the target population
  2. items must have a unique linguistic meaning
  3. items must have unique content
  4. items must be clearly related to key aspects of the construct
  5. items must have equity
  6. items should ideally be “clinically” useful
22
Q

What are some error sources?

A

-social desirability and malingering