Moving From Fact to Truth: Validity, Reliability, & Measurement

Chapter 3
Research Methods In Psychology
 Relative accuracy or correctness of a psychological
 Internal Validity- the extent to which a set of research
findings provides compelling information about
 Laboratory experiments tend to be high in internal
They control for individual differences
They isolate independent variables from potential sources of
Three Requirements for Causality
 Covariation
 the principle of correlation, i.e., the co-occurrence of or association
between two events.
 changes in one variable must correspond to changes in another
 Temporal Sequence
 establishing that a potential cause of an outcome occurs prior to the
outcome in question.
 changes in the first variable must precede changes in the second
 Eliminating Confounds
 ruling out other plausible causes of an event before concluding that
one thing is the cause of another.
 External Validity- the extent to which a set of research
findings provides an accurate description of what
typically happens in the real world.
 Generalizability of research findings- results can be
applied to other people and physical or social
 Construct Validity- the extent to which the
independent and dependent variables in a study truly
represent the abstract, hypothetical variables of
interest to the researcher.
 Whether or not the manipulated and/or measured
variables accurately reflect the variables the researcher
hoped to manipulate or measure.
 A direct reflection of the quality of the operational
 Conceptual Validity- how well a specific research
hypothesis maps onto the broader theory that it was
designed to test.
 The consistency or repeatability of a measure or
 Test-retest Reliability
 The degree to which a test continues to rank order
participants’ scores in a stable manner over time.
 Ideally assessed by correlating participants’ scores on a
measure at time 1 with their scores on the same measure
from time 2.
 Internal Consistency
 The degree to which the total set of items or
observations in a multiple-item measure behave in the
same way.
 For example, the degree to which each item in a ten-item
self-esteem inventory rank orders participants in the
same manner.
 Interrater Reliability (interobserver agreement)
 The degree to which different judges independently
agree upon a measurement, judgment or observation.
 Intrarater Reliability
 The consistency of a judge’s data across repeated scoring
of the same participant or observation (usual scored
from a videotape)
Measurement Scales
 Four types of scales
 Nominal, ordinal, interval, ratio
 Each have increasing levels of measurement and
 Each has a different mathematical attribute which is
expressed about the particular population
 Classification of objects and placing them into specific
categories (“Naming them”)
 Tends to be more of a qualitative measure rather than a
 There is no way to measure between the categories
 Nominal Scale – Religious affiliation
 Buddhist, Christian, Jewish, etc.
 Shows the magnitude of whether a particular variable
is present
 Less than, Same as, more than
 Rank ordering
 Example – College Basketball rankings
 1. Duke, 2. Kansas University, 3. Wake Forest, etc.
Interval Scale
 Has the same properties as Ordinal scales, but also has
equal distance between the units
 But does not have an absolute zero point
 Temperature is an easy example
 Movement of one degree is equal across the scale
 Contains all the properties of interval scales but does
contain an absolute zero point
 Ratios allow for multiplication of factors on the scale
 Scores may be twice or three times as high or lower than
other scores