Transcript Measurement in Psychology

Measurement in
Psychology
Measurement in
Psychology
A construct is an idea or a concept that we
wish to explore. More particularly we wish to
make measurements of about this idea.
Examples of constructs include
Intelligence
Aggression
Maturity
Depression
Measurement in Psychology

Operationalisation is the process by which
we translate the construct or idea into
something we can measure. Thus we
transform the idea of intelligence with the
use of a standardised IQ test or we may
measure depression using a depression
index according to a standard test.
Measurement in Psychology

These measures are divided into Objective
and Subjective Measures.
Measurement in Psychology

Objective quantitative measures include



Standardised tests
Physiological measures
 Heart rate
 Skin conductivity (lie detector)
 Heart rate
Behaviour counts

Eg counting aggressive acts or number of
attempts to join a conversation
Measurement in Psychology

Subjective measures include



Questionnaires
Rating scales
Open-ended questions “How did you feel?”
“What do you think?” eg Year 11 survey
Measurement in Psychology
 Describing
data
Quantitative Data – numbers
 Qualitative Data – descriptions,
pictures

Measurement in Psychology



Quantitative data
We need to sort out and describe the data
– Descriptive statistics
Use the data to make decisions or
inferences – Inferential statistics
Measurement in Psychology

We must first understand to the type of
data we are dealing with and then what
methods are best used to deal with these
data.
Measurement in Psychology

Nominal data





We classify according to category only.
Sometimes called categorical data.
Eg type of sport played
Type of car owned
Gender
Favourite type of music
Measurement in Psychology
 We
would often display these data
as a bar graph and we would
measure central tendency with the
mode. The mode is the most
popular value.
Measurement in Psychology
Car
Number of
owners
Toyundai
11
Volksburu
16
Mitsubenz
3
Measurement in Psychology
Number of owners
18
16
14
12
10
Number of owners
8
6
4
2
0
Toyundai
Volksburu
Mitsubenz
Measurement in Psychology

Ordinal Scale


Some sense of order, but the magnitude makes
no sense. Eg classifying age as child, teen-ager,
adult, elderly. There is a clear order but the
magnitude makes no sense.
We would use a bar graph and the median to
determine central tendency.
Measurement in Psychology
Age Class
Population in Hicksville
Child
120
Teen-age
180
Adult
400
Elderly
100
Total = 800
Measurement in Psychology
Popuation in Hicksville
450
400
350
300
250
Popuation in Hicksville
200
150
100
50
0
Child
Teen-age
Adult
Elderly
Measurement in Psychology




Interval Scale
Here, the magnitude of the numbers is relevant
and the intervals between the numbers can be
compared. Thus we can speak meaningfully of
the difference between 15°C and 25° as being the
same as the difference between 21°C and 31°C.
However, the zero point is not really zero and so
20°C is not half as hot as 40°C
Here we can use the mean and the median.
Measurement in Psychology




Ratio Scale – same as the interval scale,
except that the zero is now relevant and
important.
Eg distances from Adelaide to other major
centres.
Heights of buildings in a city.
Here we use the mean or the median.
Measurement in
Psychology
We can use the information about a group
of data to help determine the effect of a
treatment.
For example, imagine that a psychologist
wished to measure the effect of alcohol on
reaction times. Two groups of 28
participants are randomly selected into a
treatment group and a control group.
Measurement in
Psychology
The control group are given orange
juice and the treatment group are
given orange juice with a measured
amount of vodka added. After
“treatment” both groups are given a
reaction time test and the times are
recorded.
Measurement in
Psychology
Within each group there is considerable
variation and we need to make a
reasonable measure of the effect of the
alcohol. This variation comes about
because of two main reasons.
1. The variation in the individual’s alertness
and reaction.
2. The variation across the groups between
the individuals.
Measurement in
Psychology
With all this potential “noise” in the
system, how can we be sure that the
effect is due to the treatment, or due
to the chance factors associated with
the other sources of variation.
Measurement in
Psychology
This is where we can make use of statistics to estimate how much
of the effect is due to chance and how much can be attributed to
the treatment.
Measurement in
Psychology
Five Number Summary Alcohol group
Min.
Q1
Med
Q3
Max
30
36
42
48
54
Five Number Summary Control group
Min.
Q1
Med
Q3
Max
18
24.75
30
36.25
42
Measurement in
Psychology
Mean Alcohol group = 42.1
Standard deviation = 7.1
Mean control group = 30.5
Standard deviation = 7.2
If we have normally distributed data, we expect that
95% of the data lies within two standard deviations of
the mean, if the data are normally distributed. Thus
we can get an idea of how much variation is brought
about by chance.
Boxplot for Control Group
0
10
20
30
40
50
60
50
60
Boxplot for Alcohol Group
0
10
20
30
40
Measurement in
Psychology
In this case, we can conclude that the
difference is due to more than chance.
In psychological research, we make
use of a “test statistic” which
determines the probability that the
variation is due to chance.
Measurement in
Psychology
Thus, if the variation can be shown statistically
to have a probability of less than 0.05 (p<0.05)
to be due to chance alone, we generally accept
that the treatment has had an effect.
However, there is a 1 in 20 chance (0.05) that
the effect may have been purely due to chance.
Thus we can never be entirely sure and we
must accept that our conclusion may be wrong.
Content analysis:
We compress large amounts of data into a few
categories.
Become familiar with the transcripts – read and reread
Look for words, phrases and concepts central to the
research question.
Examine themes to find common ideas, differences
and patterns
Count how often these themes occur.
Use more than one person to do the coding.
Measurement in
Psychology
Assignment for this week
Page 38 Focus questions 1 – 12.
Measurement in
Psychology
Measurement in
Psychology
Measurement in
Psychology
Measurement in
Psychology
Measurement in
Psychology
Measurement in
Psychology