PTPR Summary for Chapters 1, 3-9, and 14 - UvA Psychology Bsc Yr 2, 26-27

Chapter 1: Psychometrics and the Importance of Psychological Measurement
Psychometrics = the science concerned with evaluating the (theoretical) attributes of psychological
tests
 Attributes of instruments are important to measure psychological attributes and processes
(especially the type of information, the reliability of data, and issues concerning the validity
of data)
o Validity = the degree to which evidence and theory support the interpretation of test
scores for proposed uses (threats include participant reactivity and observer bias)
o Construct = a theoretical psychological characteristic, attribute, process, or state that
cannot be directly observed
 Latent variable = a statistical term usually referring to the psychological
constructs driving responses to a test
o Operational definition = the procedure used to measure hypothetical constructs (or
anything)
Challenges in psychological measurement
 Complexity of concepts: difficulty identifying and capturing important aspects of
psychological attributes in a single number or score
 Composite scores = a score that is computed by summing or averaging a set of responses
made by an individual
 (Score) sensitivity = the ability of a measure to discriminate adequately between meaningful
amounts or units of the dimension that is being measured
 Participant reactivity = people’s knowledge that they are being observed can cause them to
react in ways that obscure the interpretation of the behavior that is being observed
o Demand characteristics = some people may try to understand the purpose of the
observation, changing their behavior to accommodate that purpose
o Social desirability = some participants might change their behavior to try to impress
the person doing the observation
o Malingering = some participants may change their behavior to convey a poor
impression to the person doing the measurement
 Observer (scorer) bias = a test user/researcher has a vested interest in obtaining particular
outcomes, thereby allowing bias to influence the testing/observation procedures
 (Lack of) awareness of psychometrics
Psychological test = a systematic procedure for comparing the behavior of two or more people,
where (1) the test involves a sample, (2) the samples are collected systematically, and (3) the test has
the purpose to detect differences between people
 This definition is accurate due to its generality (e.g. the definition allows for tests that are not
strictly pen-and-paper), it allows for an extended amount of information in different forms to
be produced (e.g. categorical data, numbers), and it concerns the general purpose of tests
o Differential psychology = areas of psychology that are traditionally seen as focusing
on the differences between people, often including the study of intelligence,
aptitude, and personality (experimental psychology, in contrast, focuses on the
average person)
 Interindividual differences = comparing the behavior of different
 Intraindividual differences = comparing the behavior of the same individuals
at different points in time or under different circumstances
Responses required in psychological tests
 Open-ended test = a test in which people can answer questions by saying anything they want
in response to the questions on the test
 Closed-ended test = a test that requires people to answer questions by choosing among a set
of alternative answers provided in the test
Use of psychological tests

, Criterion-referenced (domain-referenced) test = a test that is most often seen in settings in
which a decision must be made about a person’s skill level, which are based on whether the
individual’s score surpasses a specific criterion threshold
o The cutoff score is associated with a decision criterion based on some
(predetermined) standard or expected level of performance of people who might
take the test (e.g. driver’s license tests)
 Norm-referenced test = a test that is usually used to compare a person’s test score with
scores form a reference sample or a normative sample, in order to understand how the
person compares with other people
o Reference (normative) sample = the sample used to establish test norms
Timing of psychological tests
 Speeded test = a test that is time limited, in general, people who take a speeded test are not
expected to complete the entire test in the allotted time, where the test is scored by
counting the number of questions answered in the allotted time period
 Power test = a test that is not time limited, in that examinees are expected to answer all the
test questions, which are often scored also by counting the number of correct answers made
on the test
Meaning of indicators in psychological tests
 Reflective (effect) indicator = a type of indicator that is seen as being caused by a
respondent’s level of a relevant latent variable (most common) (e.g. scores on
intelligence/personality tests, as a person’s response are typically seen as caused by
intelligence/personality: the construct determines, in part, a person’s responses to the items)
o Construct -> indicators
 Formative (causal) indicator = a type of indicator that is seen as relevant to forming a score
that reflects a latent variable but that is not seen as being caused by that latent variable (e.g.
SES)
o Indicators -> construct

, Chapter 3: Differences, Consistency, and the Meaning of Test Scores
Variability (differences) = the degree to which there are differences among a set of scores
 Interindividual variability = differences between people
 Intraindividual variability = differences within a person: variability of one person’s scores at
different times or in different situations
o Distribution of scores = a set of scores on a variable of the same people at different
points in time
 Central tendency = the typical or most representative
score in a distribution of scores
 Mean = a specific statistic index representing the central tendency of
a set of scores
 Median
 Mode
 Normal distribution = a distribution of scores that,
among other more technical properties, is symmetric (a
theoretical idea which can be used as a model for
interpreting scores when the data are
very close to a normal distribution)
 Skewed distribution = a situation in
which a distribution of scores is not symmetric
 Negatively (left) skewed: skew values greater than 0
(more data on the left)
 Positively (right) skewed: skew values below 0
(more data on the right)
 The more asymmetric a distribution is, the larger its
skew value is
 Variance = degree of variability in a set of scores, in terms of the degree to which the
individual scores in a distribution differ from the mean score of that distribution: the mean of
squared deviation scores, which represents the average degree to which people differ from
each other
o Reflects variability in terms of squared deviation scores
o Variance is determined by the degree to which scores differ from each other and the
metric of the scores in the distribution
 Larger variance indicates greater variability within a distribution
 A difference in two variances can be due to dramatic difference in the
metrics of the two sets of scores
o Can never be negative (less than zero)
o Best interpretable and meaningful when the variance is put into context
 Dichotomous (binary) item = a test item for which only two outcomes are
possible
 The proportion is equivalent to the mean for a binary test item
o This means that the variance of a binary response item
depends only on the proportion (p) and q (which is 1-p), or,
in other words, the proportion of each of the two possible
responses
 Standard deviation = degree of variability in a set of scores: square root of the variance
o Reflects variability in terms of the size of raw deviation scores (more intuitive)
o Can never be negative (less than zero)
Covariability (consistency) = degree to which differences in one set of scores correspond with
differences in another set of scores (scatterplots can represent consistency quite well)
 Direction of the association (positive or negative)