Designing Effective Surveys

Conceptualizing Your Research

Conceptualize your primary research; Figure out what you need to knowIf you need information on the opinions or actions of a whole group, do a surveyIdentify key elements of the concepts you need information about

Your Survey PopulationCarefully choose and define your

survey population

The Survey PopulationAre they relevant to the problem?Are they the group affected by the problem? Or are they the group that will have to carry out the recommendations? Or are they people who have dealt successfully with this problem elsewhere?Can you get a fair sampling? How?Will you get a high response rate?

Operationalize Your Concept

Consider your audience’s needs (your survey respondents as well as target audience)Decide how to measure your concept by using reliable, valid questionsDraft closed-end questions

Survey questions should be easy to understand and hard to misinterpret!

Writing Good QuestionsUse common language & clearly defined termsUse mostly closed questionsInclude demographic questions to help identify the sampleUse precise, unambiguous languageMake sure questions are valid & reliable

Which Is the Valid Question?

How much do you like doughnuts?How many times a month would you visit a doughnut shop located within three blocks of campus?

Is This a Reliable Question?

Do you like high quality pastries?

What to AvoidAvoid double-barreled questionsAvoid leading, biased questionsAvoid loaded questionsAvoid repetitive questionsAvoid personal questions

Gathering Demographic Information

Do you expect age or gender to play a role in people’s opinions?Do you expect education level or income level to play a role?Do you expect size of household or length of employment to play a role?Do you want to compare responses from different groups?

Sequencing your Questions

Group questions that are similarPut them in a logical orderPlace demographic questions at the beginningPut any sensitive or difficult questions at the endPut any open-ended questions at the end

Types of Questionnaire Questions

Yes/No

Are you currently employed?

___ yes ___ no

True-false

Your store is located downtown.

___ True ___ false

Multiple ChoiceWhat precautions does your store

take to curtail shoplifting?___ Sensormatic tags___ Ink dye tags___ Employees circulating on the floor___ Dressing room attendants on duty___ Surveillance cameras

Agree/disagree Scale

Drug use is a problem among our employees.

___ Strongly agree___ Agree___ No opinion___ Disagree___ Strongly disagree

RankingHow much of your household’s

shopping and recreation occurs downtown?

___ Little or none___ Less than half___ About half___ More than half___ Most or all

RatingRate the following factors according to

their influence on you when you bought your home? (5 = very important and 1 = unimportant)

___ Commuting time___ The home itself___ Purchase price and financing___ Proximity to friends and family___ Schools

ChecklistFrom what sources did you finance

your business? Check all that apply.___ Bank___ Savings and loan institution___ Personal savings___ Federal or state government loan___ Other ________________________

Formulating Response Categories

Limit types of questions and response sets to no more than three Give clear-cut answer choicesMake response sets easy to navigateMake sure response categories don’t overlapMake sure responses cover all possibilities

Overlapping Response Categories

Example:How many credit hours have you

completed at IUPUI?___ 1 – 25___ 25 – 55___ 55 –85___ 85 or more

Questionnaire Title & Purpose

Give your questionnaire a topic-related titleInclude a brief explanation of the survey’s purpose (short paragraph)Include instructions for completing the surveyInclude instructions for returning the surveyAssure respondents the survey is anonymous

Revising Your Questionnaire

Have you asked everything you want answers to ?Will the answers help you make recommendations?Are your questions unbiased, clear, and to the point?

Are they logically arranged?Is the questionnaire neat & uncluttered?Is it carefully edited?Can respondents complete it quickly & without difficulty?

Be sure to test run your questionnaire!

Once questionnaires with a mistake go out, it’s too late to correct them! This has a negative effect on your credibility as a researcher.

Ways to Conduct Your Survey

Distribute & pick up questionnaires on the spot in a class, at a meeting, etc.Leave a box in a break room or other common areaSend as an attachment to each respondentConduct by telephone

Next class….How to tabulate and interpret your

findings