PYC4807 SUMMARY NOTES

PYC4807
STUDY NOTES

, lOMoARcPSD|21997160




Theme 1: Psychometric Theory


Tutorial letter 2

 The purpose of testing needs to be specified before an assessment (The combination
of psychological assessment measures that will be used).
 The characteristics of choosing the measure:
o Needs to consider what attribute, characteristic or construct it measures.
o Need to check its appropriateness for an individual, group or organisation.
o Should determine if the measure is psychometrically sound.

Norm referenced vs Criterion referenced measures

 Norm referenced implies that an individual score on the measure is interpreted by
comparing this score to the performance of people similar to himself or herself.
 A norm is defined as a measurement against which the individual score is evaluated so
that the individuals position relative to that of the normative sample can be
determined.
 The normative sample refers to the group of people for whom the test was initially
standardised during the development process.
 The test should be standardised on a representative group of the population for which
it is intended.
 Criterion referenced measures: Compare a persons knowledge or skills against a
predetermined standard

Reliability

 Reliability refers to the consistency with which a psychological test measures
whatever it measures.
 You should understand what is meant by a persons observed score, true score and
error score in the context of reliability.
 Observed score: Actual score on exam
 True score: The persons actual ability
 Error score: The difference between observed and true score

Validity

1

, lOMoARcPSD|21997160




 Validity refers to the appropriateness of the inferences made from test scores
 Validity also refers to a test that measures what it is supposed to measure.
 Content description validity (Face validity and content validity)
 Construct validity.
 Criterion validity (Concurrent validity and predictive validity)

Bias

 Bias is determined by means of objective statistical in devices and has its origin in
elements inherent to a test or in the method and process of measurement.
 Fairness is related to social values and philosophies of the assessment measure.
 Fair and ethical assessment practises should be a central to all assessment contexts.
 Psychrometric soundness does not guarantee fairness. It would be difficult to apply a
measure that is unsound fairly.
 Bias threatens the equivalence of measurement outcomes.
 Measurement bias: Systematic inaccuracy in measurement.
 Prediction bias: When group membership influences the prediction of criterion scores.
 Method bias: problems resulting from the way that an assessment is administered, the
incomparability of the samples used and the inequality produced by the specific
instruments

Chapter 3 in textbook:

Categorical measurement data can be divided into nominal and ordinal measurement levels.

 Nominal data is classified without a natural order or rank. Nominal data is used to
label variables without any quantitative value such as male or female or names of
people.
 Ordinal data has a predetermined order. Ordinal data is labelling variables with
quantitative value. For example, socio economic status, or educational level.

Continuous data is categorised into interval and ratio.

 Ratio has a zero point. Pressure has a zero point because. It’s possible for the answer
to row. For example age.
 Interval variables are variables which their central characteristic is that they can be
measured along a continuum and they have a numerical value. Example IQ.



2

, lOMoARcPSD|21997160




3.6 In textbook

 Some measures may be Norm referenced while others are criterion referenced.
 Norm referenced measures each test takers performances interpreted with reference to
a relevant standardisation sample or norm group.
 Criterion referenced. Measures, on the other hand, compare the test takers
performance to the attainment of a defined skill or content.
 A norm can be defined as a measurement against which an individuals raw score is
evaluated so that the individuals position relative to that of the normative sample can
be determined.
 The normal distribution which is referred to in Psychometrics is the standard normal
bell shaped distribution which has a mean of zero and a standard deviation of 1.

Establishing norm groups

 Small groups consist of two subgroups, the applicant pool and the incumbent
population (population affected by it).
 The similarities between the norm Group and the incumbent group are established by
means of comparative aspects such as ethnicity, gender, age and educational
background.
 For example. We want to establish a norm group as the comparative base for our pilot
selection. The Norm group should be selected from both the pool of applicants for
pilot training, (the applicant group, not yet competent)and the population of licenced
pilots, the incumbent group (competent). The Norm group should be representative of
these two groups.

Co-norming of measures

 Co norming entails the process where two or more related but different measures are
administered and standardised as a unit on the same normal group.
 Co norming can have several benefits where the same norm group is used for
comparable but different measures.
 Sample differences can. Now be overcome by using a common sample group.
 Saving time, doing two things at once

Four types of test norms

1. Developmental scales

3