Transcript Measurement slides

Basic Steps in Research
Observation
Statement of the Problem
(Research Question)
Design Study
Measurement (Collect Data)
Statistical Analysis
Interpretation (Conclusion)
• State Hypotheses
• Use/Generate a Theory
Absolute versus Relative (Comparative) Assessments
Absolute: “How many hours of TV did you watch last year?
“Is this drink sweet?” or “How sweet is this drink?”
Relative: Did you watch TV more hours than you spent reading the local paper?
“Which of these five drinks is the sweetest?”
• Generally, it is easier for people to make relative vs. absolute judgments (more
accuracy and consistency exists)
• People rarely make absolute assessments in everyday activities (most choices are
basically comparative)
Limitation with relative assessments and the instances when absolute judgments
are vital ---
Scales of Measurement
1) Nominal -- Indicates categories, classification (e.g., gender, race, yes/no)
Stats: N of cases (e.g., chi-square), mode
2) Ordinal -- Indicates relative position; greater than, less than (e.g., rank
ordering percentiles)
Stats: Median, percentiles, order statistics
1st
2nd
3rd
Does not indicate how much of an attribute one
possesses (e.g., all may be low or all may be high)
Does not indicate how far apart the people are with
respect to the attribute
3) Interval -- Indicates an absolute judgment on an attribute (equal intervals)
No absolute zero point (a score of 80 is not twice as high as a score of 40)
Stats: Mean, variance, correlation
4) Ratio -- Possesses an absolute zero point (e.g., number of units produced)
All numerical operations can be performed (add, subtract, multiply, divide)
Normal Curve
-4
-3
-2
-1
Mean
+1
Central Tendency
+2
+3
+4
Variability (Spread in scores)
a)
Mode (most frequent score)
a)
Range (lowest to highest score)
b)
Mean (average score; [EX/N])
b)
Standard Deviation
c)
Median (midpoint of scores)
c)
Variance
Computation of Standard Deviation & Variance
Test Scores
Squared
deviation
scores
Deviation scores
(scores minus the
mean
X
x
x2
10
-20
200
20
-10
100
30
0
0
40
10
100
50
20
200
EX = 150
EX2 = 1000 (Sum of the
squared deviation scores)
(EX/N) = 30 (Mean)
EX2/N = 200 (the variance or s2)
s2
= standard
deviation or s
200
Mean of the sum of the squared
deviation scores
= 14.14 (standard deviation)
Relationships Among Different Types of Test Scores in a Normal Distribution
Number of
Cases
2.14%
0.13%
2.14%
0.13%
13.59%
-4
-3
-2
-4
-3
-2
10
20
34.13%
-1
34.13%
13.59%
Mean
Test Score
+1
+2
+3
+4
-1
0
+1
+2
+3
+4
30
40
50
60
70
80
90
200
300
400
500
600
55
70
85
100
115
Z score
T score
CEEB score
700
800
Deviation IQ
(SD = 15)
4%
7%
1
2
12%
17%
20%
17%
130
145
12%
7%
4%
7
8
9
Stanine
3
4
5
6
30
40 50 60
70
Percentile
1
5
10
20
80
90
95
100
Positively Skewed
Negatively Skewed
Distribution
Distribution
40 45 55 60 70 75 80 90
Test Scores
100
40 45 55 60 70 75 80 90
Test Scores
100
6-week program
between tests
Did the program work to
increase scores?
Math
Math
English
English
Pretest
Posttest
Pretest
Posttest
55
56
33
35
64
66
35
37
44
46
43
47
33
38
36
36
28
29
20
21
63
63
60
62
48
50
40
40
38
40
31
31
46
48
52
56
47
47
64
66
%
increase
“Lying” with
numbers
100
90
80
70
60
50
40
30
20
10
0
Math
English
Some Common Designs Used in I/O Research
One-Shot Case Study
X
O = Observation or
Collection of Data
O
One-Group Pretest-Posttest Design
O
X
O
Static Group Comparison
X
O
O
X = Treatment or
Intervention
Common Designs (cont.)
Non-Equivalent Control-Group Design
O
X
O
O
O
Time-Series Design
O1
O2
O3
O4 O 5
X
O6
O7
O8
O9
O10
O6
O7
O8
O9
O10
O6
O7
O8
O9
O10
Multiple Time-Series Design
O1
O2
O3
O4 O 5
O1
O2
O3
O4 O 5
X
Fairly Uncommon Designs in I/O Research
Pretest-Posttest Control Group Design
R indicates
randomization
R
O
R
O
X
O
O
Posttest-Only Control Group Design
R
R
X
O
O
This is a graph of accident rates for a year. At
first glance, does this graph indicate anything
of importance to the organization?
50
45
40
35
30
25
20
15
10
5
0
J
F
M
A
M
J
Jul
Aug
S
O
N
D
How about now?
10
9
8
7
6
5
4
3
2
1
0
J
F
M
A
M
J
Jul
Aug
S
O
N
D
An organization reports that accidents have decrease
substantially since they began a drug testing program. In
1995, the year before drug testing, the number of accidents
was 50. In 1996, the year testing began, the amount dropped
to 40. In 1997, the year after drug testing the number of
accident dropped to 29. What do you make of this?
55
50
*
45
40
*
35
30
*
25
20
15
10
5
1995
Drug Testing
1997
65
Given the illustration below, now what
do you make of the effectiveness of the
drug testing program?
*
60
*
55
*
50
*
45
40
*
35
30
*
*
25
*
20
*
15
1992
1993
1994
1995
1996
1997
1998
1999
2000