research planning and design lecture 1 Flashcards
name the 4 broad steps in the measurement process
- conceptually defining the construct
- operationally defining the construct
- implementing the measure
- evaluating the measure
conceptual construct vs operational construct
- conceptual tells you what the concept means
- operational tells you how to measure it
construct
- is a hypothetical/ abstract concept
- for example, motivation, depression
- exisists theoretically but is not directly observable
what is operationalisation
- something we can objectively measure/ manipulate
- operationalising a construct creates measured or manipulated variables.
- helps others to replicate our study/findings
- example measures -> physiological, behavioural, other reported and self reported <- require precision
reliability vs validity
- reliability = consistency/ stability of a measure
( the true value of the variable + measurement error, getting same score over and over) - validity = does the measure, measure the construct it is supposed to measure
(the test actually measuring)
test- retest reliability
- measures the consistency of results when you repeat the same test on the same sample/person at a different point in time
- should get the same results every time
- a strong correlation between test scores would show strong relationship between two measurements.
example: 100 students take the SAT. Six months later, the same 100 students take test, look for correlations between two scores
Inter rater reliability
- is a measure of the consistency and agreement between two or more raters or observers in their assessments, judgments, or ratings of a particular phenomenon or behaviour
- if ther observations are the same, high reliability
- example, rating aggrressiveness of play behaviour.
internal reliability and split half reliability
= the measure has consistency within itself
- split half reliability = focuses on whether the different sections of a test consistently measure the same underlying construct or skill, test is divided into two halves. eg q1-5, 6-10
face validity
- does the measure make intuitive sense?
- does it measure what it’s meant to
- for example a questionairre on happiness:
i feel joyful most of the time (high face v)
i enjoy eating spicy food (low face v)
content validity
- does the measure reflect all relevant parts of the construct it aims to measure
- for example test anxiety:
does it measure both the cognitive and affective components of it?
criterion validity: concurrent and predictive
- extent to which a measurement correlates with or predicts an outcome based on a specific criterion.
-example -> the results of an IQ test can be used to predict future educational achievement.
-types:
concurrent = criterion and construct are measured simultaneously
predicitve = criterion measured after construct (predicts future outcome)
comparing content, criterion and face validity
Criterion validity is focused on the relationship between the test and a measurable outcome
construct validity examines whether a test truly measures the theoretical construct it is intended to.
Meanwhile, content validity ensures that a test covers the full range of the construct without necessarily assessing its correlation with outcomes.
convergent validity
correlates strongly with measure of same construct
- how closely a test is related to other tests that measure the same (or similar) constructs
discriminant validity
correlates less with measure of different constuct
- the extent to which a test is not related to other tests that measure different constructs.
- example to establish discriminant validity, they need to demonstrate that the self-esteem test does not correlate strongly with an unrelated construct, like intelligence.
