RM qual 4 Flashcards
typically standardised measuring instruments in surveys
what it is and how it is administered is clear
normative data often provided for comparison
purposes of surveys
Info gathering: Exploratory, Descriptive
Theory testing and building:
Explanatory, predictive.
when to design a new questionnairre?
no established tool to measure phenomena exst
when reliability or validity of established tool is in doubt
avoid jangles (diff labels of same thing)
General design principles of questionnaires
- keep it short
-make sure its readable - provide appropriate response options
what to consider about the order of a survey when constructing
- can be useful to divide qs by topic, q type
- Screening questions - is p.p eligible
- Start w ez and engaging questions
- USe funneling/branching questions if appropriate
pros of open questions
provide more detail
rich data
don’t assume
cons of open qs
longer and more difficult to complete
difficult to analyse
pros of closed qs
quick
easy to analyse
standardised responses
cons of closed qs
can impose assumptions
oversimplify complex issues
closed questions are more useful for which kinds of work
explanatory or predictive
what to consider when writing questions
don’t use negative language
don’t use double negatives
use neutral qs. don’t use emotive language
avoid using leading qs
response aquiescence
tendency to agree rather than disagree
includes both positively and negatively worded questions
to pick this up, spot contradictions of positively and negative question agreements
Dichotomous question
2 response options
simplest
e.g yes or no
Multichotomous questions
either give one response out of multiple options
or multiple responses
Likert scales
Multi-point responses
aim to ensure equal spacing of response options
agree vs disagree
things to take into consideration when using likert scales
-response acquiescence
-potential for double negatives
-verbal responses for all, or only for anchors? (e.g have strongly agree, disagree and have numbers in between OR all just verbal)
-inclusion of neutral response?
non verbal rating scale
useful with chilren or cognitively impaired indivs
-e.g. point to face that shows your answer
- sometimes labels are necessary
Ranking scales
measure relative importance of several items
e.g. rank the anxiety you experience in the following situations, 1 being highest …
Graphic rating scale
p.p mark along a continuous line which is anchored at each end
Record score by measuring where line is marked
semantic differential scales
most indirect measurement of attitude
(doesn’t assume attitudes as a cognitive belief)
Respondants indicate thoughts and feelings by marking a response on a scale between two opposite bipolar adjectives
questionnaire response rates are maximised by
- keeping em short
-simple + clear - pre-paid envolope
-send a reminder (after one or 2 weeks)
-offering an incentive
number of response options
too little will result in _____
too many will result in ____ _______
low sensitivity
low reliability
psychometrics
includes measurement of
science of measuring psychological constructs
measurements of personality, cognitive ability, attitudes, knowledge, educational attainments
Quality is assessed using two criteria
reliability
Temporal consistency (stable over time or over conditions)
Internal consistency ( extent to which tool is measuring the construct were interested in)
what is Construct validity
is the construct we are trying to measure valid? Does it actually exist itself?
how is construct validity supported
supported by cumulative research evidence over time
each piece establishing an element of reliability/validity
together, supporting evidence of construct itself
in the short term, how can construct validity be assessed
Convergent validity: correlates with tests of related constructs
Discriminant validity: doesn’t correlate with different constructs
what are psychometric tests
what are some types
standardised questionnaires/tests designed to measure particular traits or abilities
items published as an inventory
-personality inventories
- cog ability tests
- Measure of mental health status
comment on validity and reliability of psychometric tests
reliability established
validity sometimes questionable