Chapter Summary

The process of measurement in psychology begins with variables. Variables are events we learn more about by assessing how they change in relation to other variables. In general, variables can be measured, as is the case with height or time, or they can be discrete categories, such as relationship status. In psychological research, variables are often the practical assessment of abstract ideas known as psychological constructs. Psychological constructs include phenomena such as intelligence and anxiety, which are recognized as real occurrences, but may be observed directly. Operational definitions allow researchers to consider psychological constructs as they are represented and measured in their research. For example, an IQ test could be the operational definition of intelligence.

Much of psychological research is conducted using questionnaires, tests, or scales. Given the prevalence of these methods, it is important to consider the different options in scale construction, including the use of open versus closed questions (e.g., items for which there are a wide range of possible responses or a limited number, respectively). In addition, researchers may use methods to disguise the true purpose of the scale, particularly when knowledge of what is being measured might affect participants’ responses. Another way of ensuring openness and honesty among participants is the bogus pipeline technique, in which participants are ostensibly hooked up to a lie detector.

Scales can also be used to assess individuals’ attitudes about various targets. The Thurstone scale allows researchers to assign relative value to attitude items, placing individuals on a continuum based on the items with which they agree. Another incredibly common attitude measurement technique is the Likert scale, where agreement with statements is indicated on a set scale, ranging from strongly disagree to strongly agree. A semantic differential scale places opposite concepts at different ends of a spectrum, asking participants to indicate their position between these two concepts. Finally, similar to the semantic differential scale, a visual analogue scale asks participant to indicate their position on a line between two opposites. Researchers then measure the distance between the end of the line and the participant’s mark.

In the construction of all scales, certain types of items should be avoided. These include items that are overly complex, technical, ambiguous, or include double negatives, as this may lead to confusion among participants. In addition, double-barrelled items, which consider two questions at once, are problematic, as disagreement doesn’t tell us if the participant disagrees with one half, the other half, or both parts of the question. Emotive language should also be avoided, as well as leading questions, and questions that might represent an invasion of privacy. Finally, response options should allow for a wide range of responses, even weighted across a spectrum, and should be sufficiently sensitive as to capture existing variability in attitudes.

Two additional forms of testing that are common in psychology are projective tests, such as the Rorschach ink blot test, and intelligence and personality tests. Projective tests may yield very rich responses but interpretation of these responses is often subjective, which highlights the importance of inter-rater reliability in testing. Intelligence and personality tests are less subjective, but require consistent revision to reflect cultural changes over time.

When psychologists want to know something about a group of people, they could talk with everyone in the population (the group of people being studied). But this can be both time- and labour-intensive; instead, researchers work with a smaller subset, or sample, that represents that population. Sampling is a complicated process where psychologists generally seek to avoid bias. Some sampling strategies rely on mathematical probability, attempting to ensure that each person in the population has an equal chance of being selected to participate in the study. This helps us ensure sample representativeness. However, sometimes representative is not desired, and psychologists will seek out people who have experience with a particular topic of study. Regardless of the method of sampling, it is important that psychologists choose samples that best fit the goals of the research.

On additional point of consideration when measuring people is the extent to which a researcher will rely on quantitative or qualitative methods in the study. Strict quantitative research develops from a view called positivism, which suggests that the only things we can truly “know” are those things that can be measured numerically. Qualitative researchers often reject positivist views, arguing that much of our understanding of humanity comes from textual information that cannot be reduced to numbers. While “pure” quantitative and “pure” qualitative researchers exist at opposite ends of a spectrum, many researchers fall somewhere in the middle, using a combination of qualitative and quantitative methods. The value of each method is largely dependent upon one’s research question.

Additional Online Resources

“The Value of Likert scales in Measuring Attitudes of Online Learners.” By Hilary Page-Bucci (also includes information on Thurstone scales):

“Research Methods: Qualitative versus Quantitative Approaches to Gathering Evidence.” A two-part podcast:

“On Methods: What’s the difference between qualitative and quantitative approaches?” By Jean Rhodes:

“Bogus Pipeline, Bona Fide Pipeline”:


Test your knowledge of the keywords and definitions in the chapter.


Interactive Quiz for Chapter 4

Instructions: For each question, click on the radio button beside your answer. When you have completed the entire quiz, click the “Submit my answers” button at the bottom of the page to receive your results.

Question 1:

a) Using a random number generator once, then selecting every eighth student that follows
b) Selecting students along dimensions consistent with population proportions
c) Asking for volunteers
d) Posting flyers in all buildings on campus advertising an online study

Question 2:

a) Using a repeated random number generator
b) Asking for volunteers
c) Picking every third person that walks by the student centre
d) Using a random number generator once, then selecting every eighth student that follows

Question 3:

a) Surveying students who live on your floor in your dorm
b) Surveying students in different lounges on campus
c) Surveying first-year university students
d) Surveying graduate business students

Question 4:

a) operational definition
b) hypothetical construct
c) psychological construct
d) categorical variable

Question 5:

a) anything about which people can be interviewed
b) anything that can be measured mathematically
c) anything that has been standardized
d) anything that is both reliable and valid

Question 6:

a) reliability
b) validity
c) standardization
d) reification

Question 7:

a) reliable
b) valid
c) standardized
d) reified

Question 8:

a) Categorical; measured
b) Measured; categorical
c) Response; participant
d) Participant; response

Question 9:

a) Psychometric
b) Standardized
c) Reliable
d) Reified

Question 10:

a) semantic differential scale
b) visual analogue scale
c) Thurstone scale
d) Likert scale

Question 11:

a) overly descriptive language
b) double-barrelled items
c) technical terms
d) ambiguity

Question 12:

a) demand characteristics
b) response acquiescence
c) factor pressures
d) social desirability

Question 13:

a) “big Q” research
b) “little q” research
c) purely quantitative research
d) augmented quantitative research

Question 14:

a) large samples may obscure a weak design
b) large samples hide potential participant variables
c) large samples are less likely to be statistically significant
d) large samples are costly and time-consuming

Question 15:

a) evaluative
b) activity
c) potency
d) all of the above