Robert A. Cummins

1
Measurement scales and depression
Robert A. Cummins Australian Centre on Quality of
Life Deakin University
http//www.deakin.edu.au/research/acqol
2
Overview

• What are the issues under investigation?
• The problem of sub-optimal response scales
• The problem of sub-optimal depression sales
• Why are these issues important?
• Likert scales are blunt instruments
• We seem not to understand what depression
  actually is
• What are the implications?
• Our response scales may be compromising our
  measurements
• We may be misdiagnosing depression

3
Freyd, M. (1923). The graphic rating scale.
Journal of Educational Psychology, 14, 83-102.
For job interviews Does he appear neat or
slovenly in his dress?
When you have satisfied yourself on the standing
of this person in the trait on which you are
rating him, place a check at the appropriate
point on the horizontal line. You do not have
to place your check directly above a descriptive
phrase. You may place your check at any point on
the line. (p.88).
then standardized to 0-10
4
Rensis Likert
Head Survey Research Center University of
Michigan

• 1903 - 1981

Likert, R. (1932) A technique for the measurement
of attitudes. Archives of Psychology, No.14, New
York.
Why only five levels of choice?
5
Problem 1

1. People can make more than five points of
   discrimination. They are therefore blunt
   instruments, not capturing the full extent of
   discrimination

6
Increasing the number of choice points above 5
increases scale sensitivity

• Diefenbach, M.A., Weinstein, N.D., OReilly, J.
  (1993). Scales for assessing perceptions of
  health hazard susceptibility. Health Education
  Research, 8, 181-192.
• Russell, C., Bobko, P. (1992). Moderated
  regression analysis and Likert scales Too coarse
  for comfort. Journal of Applied Psychology, 77,
  336-342.
• Jaeschke, R., Guyatt, G.H. (1990). How to
  develop and validate a new quality of life
  instrument. In B. Spilker (Ed.) Quality of life
  assessment in clinical trials (pp.47-57). New
  York Raven Press.

7
Problems with Likert scales

• People can make more than five points of
  discrimination. They are therefore blunt
  instruments
• 2. The number of choice points cannot easily be
  expanded because we do not have the necessary
  adjectives

8
Roy Morgan Research (1993)
Roy Morgan Research (1993). International values
audit, 22/23 May. Melbourne Roy Morgan
Research Centre.
9
Problems with Likert scales

• People can make more than five points of
  discrimination. They are therefore blunt
  instruments
• 2. The number of choice points cannot easily be
  expanded because we do not have the necessary
  adjectives
• 3. The psychometric distance between the named
  adjectives does not accord with the interval
  nature of the scale

10

• Ware and Gandek (1994) used the Thurstone method
  of equal-appearing intervals to calculate the
  following distances between category labels used
  in the SF-36

Ware, J. E., Gandek, B. (1994) The SF-36 Health
Survey Development and use in mental health
research and the IQOLA project. International
Journal of Mental Health, 23, 49-73.
11
Louis Leon Thurstone (1887 -1955)
Dept Psychology University of Chicago
Jones, L.V., Thurstone, L.L. (1955) The
psychophysics of semantics An experimental
investigation. The Journal of Applied
Psychology, 39(1), 31-36.
12
(No Transcript)
13
How satisfied are you with your life as a whole?
Can people reliably use 11-points of
discrimination?
N30,000
14
The relationship between SWB and depression
Critical issue 2

• Can the Subjective Wellbeing, or Positive Affect,
  be used as a measure of depression?

15

• Dominant Source of SWB Control

Defensive range
80
a
70
b
c
Strong homeostatic defense
Lower Threshold
Strength of challenging agent
16
Theoretical proposition
Loss of positive wellbeing depression

• Positive wellbeing is controlled by a homeostatic
  process
• Homeostatic defeat means positive wellbeing is
  lost
• Depression is the loss of positive wellbeing
• The measurement of positive wellbeing should be
  THE measure of depression

17
Is this idea consistent with (DSM-IV) ?

• Symptoms of depression include the following
• (a) depressed mood (such as feelings of sadness
  or emptiness)
• (b) reduced interest in activities that used to
  be enjoyed, sleep disturbances (either not being
  able to sleep well or sleeping to much)
• (c) loss of energy or a significant reduction in
  energy level
• (d) difficulty concentrating, holding a
  conversation, paying attention, or making
  decisions that used to be made fairly easily
• (e) suicidal thoughts or intentions.

18
Depression is---

• (a) Loss of positive affect due to homeostatic
  failure
• reduced interest
• loss of energy
• (d) difficulty concentrating
• suicidal thoughts

These are just the consequential symptoms caused
by the loss of positive affect
19
How do we establish that loss of positive
wellbeing depression?
How do the measures of depression and SWB relate
to each other?
r .7
20
How do the distributions of population incidence
match?
Incidence of depression in Australia
Commonwealth Department of Health and Aged Care (2000) 5.8
1997 National Survey of Mental Health and Wellbeing 6.0
21
The relationship between SWB and depression
(symptoms) follows the theoretical pattern
prescribed by homeostasis.
22
Conclusions

• 11-point end-defined scales are superior to
  Likert scales
• (b) Depression should be defined, and measured,
  as a loss of positive affect.