lecture 7 Flashcards
A poll consists of
one multiple choice question
Surveys consist of
A large number of questions across a wide range of question types
Example of a poll
A company has committed 1% of its profits to charity. However, it wants its employees to decide where they should spend it. In such a case, it can create a poll for its employees and make it live on the company’s intranet portal, asking employees for their preferences. Free online polls can for example be created here.
Example of a survey:
A company wants to learn about its customers’ experiences to improve the quality of its products and minimize future negative feedback. In such a case, it can create a customer satisfaction survey, asking a number of questions about customers’ overall satisfaction and their experience with several features of the product.
Polls only allow for
Descriptive research
Descriptive reserach
Research done to describe reality
Eg, you may need to know the frequency of a phenomenon before beginning to theorize what is associated with it. however, facts alone do not explain why things are what they are.
Explanatory research
Research done to test theories of associations
When to use survey research
Survey research is not always the best approach to address a research question.
When should we use survey research
Survey research is particularly useful to find out peoples (consumers, managers, investors,mmm) perceptions, beliefs, annd attitudes. For example, do people believe that investing in art is good or bad? do they have a positive attitude towards a product or not? How satisfied are they with a service provided
For every variable in their conceptual model, survey researchers need to decide whether to use:
An open ended or a clost ended question
A single item or a multi item measure
An off the shelf measure or an own developed measure
Open ended questions
Allow respendents to answer any way they like
Drawback is that they are relatively difficult and time consumer for researchers to code and analyze
Closed-ended questions
Offer a number of answering options from which the respondent is instructed to pick. Closed-ended questions are more efficient to analyze (especially for large groups of respondents)
Response formats for closed-ended questions
Ranking questions: where the respondent is asked to rank order something
Rating questions: where the respondent is asked to rate a statement
Categorical questions: where the respondents answer can fit only one category
Comparative scales:
Paired comparison
rank ordering
Constant sum (e.g. divide 100 points among the following brands, indicating your preference for personal use)
Non-comparative scales (rating scales)
Continuous rating scale (how would you rate de bijenkorf as a department store 0-100
Likert scale (agree or disagree with a statement 1 not at all 7 very much)
semantic differentials:
What do you think of the apple iphone
Good/bad
powerful/weak
Modern/old fashioned
response categories for categorial scales
Should be mutually exclusive (only 1 answer applies)
Should be collectively exhaustive: the answer possibilities cover the entire realm of possibilities
single item measure
Consists of a single questions
Multi item measure
Consists of two or more questions
single item measure
When concrete singular object/attribute:
What is your marital status
What is your profession
Multi item measures
Use off-the-shelf scales
Develop your own scale
off-the-shelf scales (advantages/disadvantages)
Advantages: known and “good” validity and reliability
Comparability of results
Low costs
Disadvantages: Not tailored to your exact research need
Require translation if in different language (source of error)
Double barreled questions
A questions that lends itsself to different responses of its sub parts
E.g. How is the taste and appearance of your pancakes
1. might like the look but haste the taste
Ambiguous questions
a questions that has no specific query
Leading questions
Phrased in a way that they lead the respondent to give the answer you want
Developing questions (loaded questions)
A questions that has an alterior motive
Double negatives
Questions containing two negative words
Survey modes
Interview assisted
-Telephone or personal
Self administered
-Mail or electronic
Survey mode decision (factors that play a role)
1) measurement
-interactivity, multi-media, interviewer presence or self-administration
2) Representation
-Coverage quality
-Sampling control
-Response rate
3) Economics
-Sample size
-Questionnaire size
-Survey speed
-Survey costs
mixed mode surveys
Combine different ways (modes) of collecting data for a single reserach project. you may use mixed mode survey designs to address problems associated with the under-coverage of key groups of interest or to improve response rates
Some mixed-mode surveys can also be conducted at a lower cost than single-mode surveys