Transcript Document

Chapter 8
Data Collection:
Primary Data
© 2005 Thomson/South-Western
1
Figure 1: Choices when Collecting Primary Data
Communication
Degree of
Structure
Structured
Unstructured
Degree of
Disguise
Disguised
Undisguised
Method of
Administration
Observation
Personal Interview
Telephone Interview
Mail Questionnaire
Degree of
Structure
Structured
Unstructured
Degree of
Disguise
Disguised
Undisguised
Setting
Method of
Administration
Natural
Contrived
Human
Mechanical
2
Figure 2: Structure and Disguise in Communication Methods
Structured
Undisguised
Disguised
Unstructured
Typical Questionnaire
(very frequently used)
Interviews
Open-Ended Questions
least used
Motivation Research
Word Association
Sentence Completion
Story Telling
3
Primary Data: Overview
•
Types of Primary Data:
–
–
–
–
–
–
–
•
Demographic / Socioeconomic Characteristics
Psychological / Lifestyle Characteristics
Attitudes / Opinions
Awareness / Knowledge
Intentions
Motivation
Behavior
Qualities of Primary Data:
– Versatility
– Business logistics: speed and cost
– Data quality: objectivity and accuracy
•
Two Classes of Primary Data
– Observation
– Communication
4
Behavior Checklist
Purchase
Use
Behavior
Behavior
What
How Much
How
Where
When
Who
5
Primary Communication Methods of Data Collection: Advantages and Disadvantages
Advantages
Disadvantages
Mail
* May be only method able to reach
respondent
* Sampling frame easily developed
when
mailing lists are available
*Not subject to interviewer bias
*Very little control in securing response
from specific individual
*Cannot secure response from
illiterates
*Cannot control speed of response;
long
response time
*Respondents work at their own
pace
*Researcher cannot explain ambiguous
*Assures anonymity of respondent
*Does not allow probing with openended
*Wide distribution possible
*Best for personal, sensitive
questions
*Generally least expensive
questions
questions
*Difficult to change sequence of
questions
*Sequence bias; respondents can view
6
entire
Primary Communication Methods of Data Collection: Advantages and Disadvantages
Advantages
Telephon
e
*Relatively
low cost
*Wide distribution possible
*Interviewer supervision is strong; less
interviewer bias
*Relatively strong response rates
(much higher than mail surveys)
*One of quickest methods of data
collection
*More difficult than with personal
interviews to determine that
appropriate respondent is being
interviewed
*Less difficulty and cost in handling
“call backs” than in-home interviewer
Disadvantages
*Difficult to establish representative
sampling frame due to unlisted
numbers
*Cannot use visual aids
*More difficult to establish rapport
over the telephone than in person
*Does not handle long interview well
in most cases
*Subject to some degree of
interviewer
bias (but much less than with
personal
interview)
*Allows easy use of computer support
*Sequence of questions is easily
7
Primary Communication Methods of Data Collection: Advantages and Disadvantages
Advantages
Disadvantages
In-Home Personal Interview
* Probably highest response rate
*Generally narrow distribution
*Best for getting response from specific,
*Interviewer supervision and control
difficult to maintain
identified person
*Allows use of any type of question/
questionnaire
*Sequencing of questions is easily
changed
*Often difficult to identify individuals to
include in sampling frame
*Generally most expensive method of
administration
*Allows probing of open-ended questions
*Costly to revisit “not-at-homes”
*Allows clarification of ambiguous
questions
*Relatively slow method of
administration
Mall easy
Intercept
*Permits
use of visuals
*Subject to interviewer bias
Same Advantages as In-Home
Interview, PLUS:
*Less expensive than in-home interview
* Sample control is more difficult than
with in-home personal interview in
terms of identifying a representative
sample
*Much better interviewer supervision and
control than in-home interview
*Interviews typically need to be shorter
than in-home interview
* Relatively short project completion time
8
Primary Communication Methods of Data Collection: Advantages and Disadvantages
Advantages
Disadvantages
Internet
* Very fast turnaround
* Very inexpensive
*Fairly versatile (e.g., can show
print ads, beauty shots of
products, can play music, video,
in addition to survey)
*Survey responses automatically
entered into a data file
*Sample is still not
“representative” of general
consumer markets (U.S. or
worldwide)
*Response rates are dropping as
novelty declines
*Respondents may have concerns
with privacy
*International sample possible
9
Figure 3: Comparing Methods of Survey Administration
HIGHLY
VERSATILE
LIMITED
phone
mail
fax
email
web
personal
(e.g., mall, home)
LOW
COST
HI COST
personal
phone
fax
mail
web
email
FAST
TURNAROUND
SLOW
mail
personal
fax
web
phone
email
10
Comparing Methods of Administering Questionnaires
Technique:
Doorto-Door
Mall
Phone
Response Rates
15%
29%
75%
Mail
Fax
Email
Web
35-63%
25%
8-37%
26%
%Bad Addresses
0-19%
41%
19-20%
24%
Response Time in Days (Mean)
13-18
9
4-6
7
Days (Median)
12
12
2
5
#Days to Receive 45%
Responses
13
1
#Days to Receive 80%
Responses
28
9
Fixed Costs
$59
$57
$57
$57
Unit Cost
$1.56
$0.56
$0.01
$0.01
Variable Costs (200 surveys
sent)
$312
$112
$2
$2
Total Cost
$371
$169
$59
$59
11
Table 7: Comparing Mall and E-Panels:
Demographics are
starting to look similar:
Responses to marketing
questions show:
1) strong consistency
(reliability measured as
the correlation between
responses over two
survey occasions) and
2) correlations between
sample methodologies
suggests modality is not
critical and should not
create bias
household size
average age
employed
white
male
college
mall
tests
2.8
40.5
71%
86%
20%
40%
internet
tests
2.9
39.2
72%
88%
21%
43%
panel
members
3.0
37.2
69%
89%
15%
46%
Correlation between Responses:
purchase intent
frequency
liking
price / value
mall vs.
internet
.86
.94
.85
.90
internet
test/retest reliability
.94
.97
.91
.99
12
Table 7, continued
Other Comparative Observations
Time survey took to administer
Internet
12.5
Phone
19.4 minutes
Upon completion, would respondent
participate in future studies?
35% yes
26% yes
More experienced Internet Users
x
Used rating scale extreme “endpoints”
more frequently
x
Jeff Miller and Alan Hogg “Internet vs. Telephone Data Collection” Burke White Paper series 2 (4)
(www.burke.com). Also see Ashok Ranchhod and Fan Zhou “Comparing Respondents of E-Mail
and Mail Surveys,” Marketing Intelligence & Planning 19 (2001), 254.
13