Types of Measures

Interval / Continuous
Every possible value included
Managing and Manipulating Survey Data:
A Beginners Guide Ordinal
All values can be placed above or below one
Chase Harrison
another
Department of Government
Harvard University
Nominal
Unique discrete categories

Types of Statistics Types of Survey Questions

Mean (average) Open-Ended
Median Ordered Scales
Percentile Discrete (yes/no)
Percentage

Coding Assigns Categories to the

Open Ended Questions
What do you think is the most important
problem facing the country at the present
time?
Data: Well, its mostly about
unemployment. A lot of people dont have
jobs. The war in Iraq is really bad too.
Responses
Jobs/Unemployment/Lack of jobs/Looking
for work
The war/Iraq/Soldiers not coming home
Terrorism/Homeland security/Protection
from terrorists

1
 The Next Step is Assigning
Questions to Ask
Numbers to Data
What about responses like: Code definitions:
The economy Jobs/Unemployment/Lack of jobs/Looking for work
Terrorists blowing people up in Baghdad The war/Iraq/Soldiers not coming home
Bushs silly war Terrorism/Homeland security/Protection from terrorists
The Democrats not supporting the war
Code Labels can be shorthand descriptions of the fuller data
Jobs/Unemployment/Lack of jobs/Looking for work Unemployment
The war/Iraq/Soldiers not coming home The Iraq War
Homeland Security
Terrorism/Homeland security/Protection from terrorists

Numbers often make nominal categories easier to manipulate

Depending on your hypotheses, you might combine new responses into
1 Unemployment
existing ones or create new categories
2 The Iraq war
3 Homeland Security

Sometimes Open-Ended Questions

Closed-End Questions
can Be Pre-Coded
What house (at Harvard) do you live in?
What is your concentration?
What state are you from?

Often its easier to ask these open-end than to

provide all categories

Example: Coding Turns this Into Numbers

How likely do you think there will be a major
terrorist attack in the next six months? 1 Extremely likely
2 Very likely
3 Somewhat likely
Extremely likely
4 Not very likely
Very likely
5 Not likely at all
Somewhat likely
99 No response
Not very likely
Not likely at all

2
 Data Data:
Percent without
Frequency Frequency Percent nonresponse
1 Extremely likely 59 1 Extremely likely 59 8% 8%
2 Very likely 146 2 Very likely 146 19% 20%
3 Somewhat likely 323 43% 44%
3 Somewhat likely 323
4 Not very likely 167 22% 23%
4 Not very likely 167 5 Not at all likely 39 5% 5%
5 Not at all likely 39 99 Refused 18 2%
99 Refused 18
Total 752
Total 752

Questions Discrete Categories

Should your percentages include or Please indicate which of the following
exclude people who say dont know from extracurricular activities you participate
the base? in
Should your percentages include or Political Organizations
Cultural/Ethnic clubs
exclude people who didnt answer the
Performing Arts Groups
question from the base?
Varsity Athletics
Intramural Athletics
Other [Please Specify]

Coding Discrete Categories Numbers

Each item can serve as its own measure Last week, how many times did you eat fish?
0 [Never/Didnt eat fish]
Political Organizations (Yes/No/No Response) 1
Cultural/Ethnic clubs (Yes/No/No Response) 2
Performing Arts (Yes/No/No Response) 3
4
5
One convention is to code these: 6
1=Yes 7
0 = No 8 or more
99 = Missing Dont know

3
 Numbers Considerations
In a data file, you may give a numeric code for Dont know or
Can be asked directly or in groupings Missing
Think about special cases: Common codes:

0 / None 98 = Dont know

Doesnt apply 99 = Refused / NA

Negative codes are sometimes more useful
Maximum
Be sure not to include numbered codes for non-numeric responses
in averages
If a large portion of respondents answer the maximum category (e.g.
8 or more), means and other statistics might have problems

Using Categories for Numbers Using Categories for Numbers

Approximately how many hours per week do you spend Please estimate your total household income in 2006,
on academic work, outside of lectures or sections? before taxes

0 - 10 hours Less than $25,000

11 20 hours Between $25,000 and $50,000
21 30 hours Between $50,000 and $75,000
31 40 hours Between $75,000 and $125,000
41+ hours Between $125,000 and $200,000
Greater than $200,000

Coding Categorized Numbers Attitude Scales

1 0 - 10 hours
2 11 20 hours Please rate the following aspects of advising in your high school by placing one check for each aspect.

3 21 30 hours
4 31 40 hours
Very
5 41+ hours Advising Aspect Very Dissatisfied Dissatisfied Neutral Satisfied Satisfied

99 No response
Frequency of contact with advisor

1 Less than $25,000

2 Between $25,000 and $50,000
Advisor's knowledge of your schedule
3 Between $50,000 and $75,000 and requirements

4 Between $75,000 and $125,000

5 Between $125,000 and $200,000 General availability of your advisor as a
6 Greater than $200,000 resource

99 No Response
Adequate advisor to student ratio

4
 Attitude Scales Analyzing Attitude Questions
1 Very Dissatisfied Percentage
2 Somewhat Dissatisfied One category
3 Neutral Two Collapsed Categories
4 Somewhat Satisfied
5 Very Satisfied Numeric
97 Doesnt apply Mean number
98 Dont know Realize this is an ordinal mean
99 Refused/No Answer Numeric scale

Creating Scales from Multiple

Examples
Questions
Possible to create scales from multiple Please indicate which of the following
questions extracurricular activities you participate
Can measure activities or attitudes in
Political Organizations
Often treated as interval data
Cultural/Ethnic clubs
Mean or Median can be reported Performing Arts Groups
Sometimes scaled to 1, 10, or 100 Varsity Athletics
Reliability of scale should, ideally, be Intramural Athletics

checkedcan be checked

Additive Scale Example of Attitude Scale

Respondent X.
1 Political Organizations
0 Cultural/Ethnic clubs
1 Performing Arts Groups
0 Varsity Athletics
0 Intramural Athletics

Additive Scale: Number of activities

Scale for Respondent X: 2
This has meaning
Seniors participated in an average of two different
extracurricular activities, compared to freshmen, who
participated in four activities.

5
 Attitude Scales Data for Respondent X
Please rate the following aspects of advising in your high school by placing one check for each aspect.
Contact 3 [Neutral]
Knowledge 5 [Very Satisfied]
Very
Advising Aspect Very Dissatisfied Dissatisfied Neutral Satisfied Satisfied
Availability 1 [Very Dissatisfied]
Frequency of contact with advisor
Ratio 3 [Neutral]
Advisor's knowledge of your schedule
and requirements

General availability of your advisor as a

resource

Adequate advisor to student ratio

Interpreting Data Interpreting Data

Individual Items Students were more likely to be satisfied with their
advisors knowledge of requirements, compared to
Single Items their advisors availability. Eighty-five percent (85%)
Collapsed of students were satisfied with their advisors level of
knowledge, with 45% reporting they were extremely
satisfied. This is in contrast with compared to only 65
percent of private school students. However, only 42
percent of students were satisfied with their advisors
availability, with only 20% being very satisfied.

Creating an Attitude Scale Scale Data for Respondent X

Make sure all answers are in same Contact 2 [Neutral]
direction: Knowledge 4 [Very Satisfied]
i.e. Positive attributes are scaled HIGH and Availability 0 [Very Dissatisfied]
negative attributes are scaled LOW Ratio 2 [Neutral]
Often useful to make 0 the lowest category
Add questions together Unstandardized Scale: 8
Standardize scale so that 0 is minimum and Standardized Scale: 50
100 is maximum (scale is arbitrary (Minimum = 0; Maximum = 16; Respondent=8)
convention)

6
 Interpretation of Scale Satisfaction with Advising by Type of School

0.9

Best not to treat as if it is real 0.8

0.75
0.80

0.7

Useful to compare means across 0.6

respondents

Satisfaction
0.5
0.43

0.4

0.3

0.2

0.1

0
Public High School Charter High School Private High School
Type of School