Constructs and Variables

1
Constructs and Variables

• S. Kathleen Kitao
• Kenji Kitao
• Keywordsconstruct, variable, operationalization,
  operational definition

2

• will discuss
• Relationship between
• actual characteristic the researcher is looking
  at
• how that characteristic is measured

3
Differences between Constructs and Variables

• A variable is essentially what we can observe or
  measure of a characteristic or ability.
• A construct is the underlying characteristic or
  ability itself.

4

• Measuring constructs
• Some constructs, such as height or weight can be
  measured directly.
• Other constructs, such as closeness of a
  friendship, cannot be measure directly.
• Researchers use questionnaires, interviews, or
  observation instead.

5

• Example of questionnaire measuring closeness of
  friendship
• Ask participants to agree or disagree with
  statements like these
• 1. I spend as much time with my friend as
  possible.
• 2. If my friend needed 100, and I had the money,
  I would give it to her.
• 3. I can trust my friend with a secret.
• These measure such aspects of closeness of a
  friendship as time spent together, willingness to
  help, and trust.

6

• It is important to keep in mind, however, that
  though the results of such a questionnaire are
  intended to reflect the closeness of a
  friendship, the results are not themselves the
  closeness of a friendship.
• In doing or evaluating research, you must think
  about how well the measures used for the
  variables actually reflect the underlying
  construct they are intended to reflect.

7
Operationalization

• Operationalization is the link between the
  construct and the way it is measured the
  operational definition of a variable is the way
  that it is measured or quantified.
• For example, English proficiency might be
  operationalized as
• a TOEFL score
• a rating of a person's ability to carry on a
  conversation in English under certain conditions
• the results of a reading test.

8

• Requirements of an Operational Definition
• 1. It must be conceptually linked to the
  construct that it is intended to represent, that
  is, it must measure the characteristics that the
  concept actually has.
• In the questions about the closeness of a
  friendship above, the three questions could be
  intended to measure time spent together,
  willingness to help, and trust.

9

• It is assumed that these are some of the
  characteristics that a close friendship would
  have.
• In measuring closeness of friendships, you should
  think about the characteristics of friendship and
  ask questions about those characteristics.
• In addition, the questions should not measure any
  characteristics that should not be included in
  the construct.
• For example, the question, "My friend is rich"
  would not be an appropriate as part of the above
  questionnaire, if the amount of money the friend
  has is not a characteristic of closeness of
  friendship.

10

• If you are measuring English proficiency, you
  need to think about what kind of proficiency you
  want to measure.
• If your hypothesis is that the higher speaking
  proficiency is, the better chance a nonnative
  speaker has of forming friendships with
  Americans.

11

• For the purpose of your research, you need to
  measure speaking. What is the best way of
  measuring speaking?
• You could ask subjects of the research to prepare
  a speech. But that type of speaking is not
  related to forming friendships.
• In this case, a better way to measure speaking
  ability would be to have the subject carry on a
  conversation and rate speaking ability based on
  that conversation.

12

• 2. Secondly, an operational definition must be
  unique to that concept, that is, it must measure
  that concept and no other.
• For example, a test of listening proficiency that
  requires some reading may also be measuring
  reading proficiency. That is, the score for
  listening proficiency may be influenced by
  ability to read.
• In that case, the operational definition includes
  a concept that the researcher did not intend to
  measure.

13

• The Importance of Operational Definitions
• Operational definitions are important for three
  reasons.
• 1. A clear operational definition allows
  researchers to be consistent throughout the
  research.
• 2. it allows readers of the research to clearly
  understand the findings.
• 3. It allows other researchers to do the same
  research to see if they get the same results.

14

• Clear definitions of constructs are important.
• In considering the relationship between variables
  and constructs when you read a research paper or
  do research yourself, it is necessary to consider
  the definitions that the researcher uses for the
  construct, if they are specified.

15

• Whether a variable truly reflects a construct is,
  in part, dependent on how that construct is
  defined.
• For example, if the definition of English
  proficiency is related to fluency in speaking
  English, then a TOEFL score is not a good
  operationalization, because the TOEFL emphasizes
  a knowledge of grammar, reading comprehension,
  and other factors that are not be related to the
  ability to fluency in speaking.

16
Conclusion

• Social scientists study many phenomena that
  cannot be measured directly.
• The actual phenomenon is called a construct the
  results of the measurement of that construct is
  called a variable.
• One of the most important issues in social
  science is the relationship between constructs
  and variables, and whether a particular variable
  accurately represents a construct.