Survey design workshop

11,647 views

Published on

This purpose of this workshop is to facilitate novice participants through the typical steps recommended for the design of a research survey (or questionnaire). The focus in on the design and development of surveys; it is not about data analysis.

Published in: Education, Technology, Business

Survey design workshop

  1. 1. Survey Design Workshop Inter-University Research Workshop Program Dr. James Neill Centre for Applied Psychology University of Canberra 11 February, 2009
  2. 2. Outline <ul><li>Workshop Objectives
  3. 3. Introductions
  4. 4. Logins & Resources
  5. 5. Research methods
  6. 6. Questionnaire design
  7. 7. Implementation
  8. 8. Conclusion & Evaluation </li></ul>
  9. 9. Workshop Objectives <ul><li>To examine issues surrounding questionnaire design including question style, response formats, layout, and pilot testing
  10. 10. To consider implementation issues (such as sampling)
  11. 11. To demonstrate the importance of rigour in planning, developing, and implementing research questionnaires </li></ul>
  12. 12. Learning Outcomes <ul><li>Understanding common methods for the design and implementation of survey-based research and the pros and cons of each method (e.g., f2f interview, mail survey, internet survey).
  13. 13. Understanding the importance of a rigorous, step-by-step process for the development of instrumentation
  14. 14. Planning, drafting, and/or revising of an initial draft (pilot) survey. </li></ul>
  15. 15. Resources <ul><li>Survey Design Workshop Notes (Wikiversity)
  16. 16. Readings </li></ul>Books about designing surveys and survey research... Look for texts like these in your libraries.
  17. 17. Introductions Introductions
  18. 18. What is a survey? A “standardised” stimulus A measuring instrument
  19. 19. What is a survey? A way of converting “fuzzy stuff” into hard data for analysis
  20. 20. Purposes of survey research <ul><li>Information gathering & describing </li><ul><li>e.g., polls, attitudes, demographics </li></ul><li>Theory-building & testing </li><ul><li>Explanatory, e.g., why?
  21. 21. Predictive, e.g., what is likely to happen? </li></ul><li>Often survey research does some of both. </li></ul>
  22. 22. Types of Surveys (Research Purposes) Descriptive <ul><li>Collects basic descriptive data/statistics e.g., consumer profiles…(age, gender) </li></ul>Explanatory <ul><li>Examine underlying data patterns
  23. 23. Linked to a hypothesis/research objective </li></ul>
  24. 24. Types of Surveys (Research Purposes) Predictive <ul><li>What happens if…
  25. 25. Useful for marketing or assessing consumer behavior
  26. 26. Honours-MA-Ph.D survey research </li></ul>
  27. 27. Research Paradigms Phenomenological Positivist <ul><li>Usually small # of participants
  28. 28. More about opinions & experiences
  29. 29. Focused on ethnographic style of research
  30. 30. Usually associated with qualitative data </li></ul><ul><li>Deductive
  31. 31. Seeks to explain causal relationships between variables
  32. 32. Usually uses quantitative data
  33. 33. Follows a structured approach or methodology </li></ul>
  34. 34. Quantitative Research Methods <ul><li>Advantages </li><ul><li>Larger sample sizes  greater confidence in results
  35. 35. Results can be representative of target population
  36. 36. Data can be readily summarised using computers </li></ul></ul>
  37. 37. Quantitative Research Methods <ul><li>Disadvantages </li><ul><li>Impersonal (don’t get background context)
  38. 38. Large samples needed
  39. 39. Samples may not be representative
  40. 40. Cost and time involved </li></ul></ul>
  41. 41. Types of Questionnaires Self - administered Interview - administered Postal questionnaire Delivery and collection questionnaire Telephone survey Face to face structured interview Web-based
  42. 43. Alreck and Settle (1995:26)
  43. 44. Questionnaire Planning/Design 1. Formulate Generic Questionnaire 2. Expand the Questionnaire Based on study objectives Turn into separate sections Question styles & types 3. Finalise Questionnaire -Pre-test/pilot test -Several drafts needed Placement & Funnel Qs
  44. 45. Formulate Generic Questionnaire <ul><ul><li>Turn objectives into sections of the survey
  45. 46. Ensure all questions relate to research objectives
  46. 47. For explanatory objectives or hypotheses ensure both dependent and independent variables exist </li></ul></ul>
  47. 48. Cover Letter / Ethics Statement Outline details of research project <ul><li>Purpose
  48. 49. What's involved?
  49. 50. Explain any risks/costs/rewards
  50. 51. Contact details
  51. 52. Human Ethics approval #
  52. 53. How is consent given/not give?
  53. 54. How to return?
  54. 55. Can choose not to continue anytime </li></ul>
  55. 56. Instructions <ul><li>Provides consistency - helps to ensure standard conditions across different administrations
  56. 57. Explain how to do the survey in a user-friendly manner
  57. 58. Example: Life Effectiveness Questionnaire </li></ul>
  58. 59. Expanding the Survey
  59. 60. Screening <ul><li>Does the participant qualify for the survey? (esp. for internet surveys)
  60. 61. Ask screening questions first, rather than later
  61. 62. Use branching if there are conditional questions </li></ul>
  62. 63. Flow/Structure <ul><li>Logical order of questions (use sections)
  63. 64. Use funnel questions to move respondents through survey
  64. 65. Start off with easy to answer and engaging questions
  65. 66. More controversial questions in middle section </li></ul>
  66. 67. Personal Information <ul><li>Put personal information at end of survey (if required), otherwise it may put off respondents
  67. 68. Consider response format e.g, Income in categories or ranges </li></ul>
  68. 69. Personal Information <ul><li>More likely to respond to personal questions for anonymous or mail surveys as opposed to face to face or telephone
  69. 70. Show cards for responses may help for face to face interviews </li></ul>
  70. 71. Types of Questions Be able to justify and defend your choices ...
  71. 72. Types of Questions
  72. 73. Open-ended Questions <ul><li>Rich information can be gathered
  73. 74. Useful for descriptive, exploratory work
  74. 75. Difficult and subjective to analyse
  75. 76. Time consuming </li></ul>
  76. 77. Closed-ended Questions <ul><li>Important information may be lost forever
  77. 78. Useful for hypothesis testing
  78. 79. Easy and objective to analyse
  79. 80. Time-efficient </li></ul>
  80. 81. Open-ended Question Examples <ul><li>What are the main issues you are currently facing in your life?
  81. 82. How many hours did you spend studying this week? _________ </li></ul>
  82. 83. Closed-ended Question Types <ul><li>Dichotomous questions
  83. 84. Multichotomous questions
  84. 85. The list (multiple response)
  85. 86. Ranking
  86. 87. Likert Scale
  87. 88. Graphical Scale
  88. 89. Semantic Differential
  89. 90. Non-verbal (Idiographic) </li></ul>
  90. 91. Dichotomous Simple Yes / No response e.g., Excluding this trip, have you visited Canberra in the previous five years? __ Yes __ No
  91. 92. Multichotomous More than two possible answers e.g., What type of attractions in your current trip to Canberra most appeal to you? __ historic buildings __ museum/art galleries __ parks and gardens
  92. 93. Multichotomous How many hours did you spend studying this week? __ less than 5 hours __ > 5 to 10 hours __ > 10 to 20 hours __ more than 20 hours
  93. 94. Frequency Scale Over the past month, how often have you argued with your intimate partner? 1. All the time 2. Fairly often 3. Occasionally 4. Never 5. Doesn’t apply to me at the moment
  94. 95. The List (Multiple Response) Provides a list of answers for respondents to choose from e.g., Tick any words or phrases that describe your perception of Canberra as a travel destination: __ Exciting __ Important __ Boring __ Enjoyable __ Interesting __ Historical
  95. 96. The List (Muliple-Response) What are the main issues you are currently facing in your life? (tick all that apply) __ financial __ physical/health __ academic __ employment/unemployment __ intimate relations __ social relations __ other (please specify)
  96. 97. Ranking Helps to measure the relative importance of several items Rank the importance of these reasons for taking a holiday to Canberra (from 1 (most) to 4 (least)): __ to visit friends and relatives __ for business __ for educational purposes __ for holiday/ sightseeing
  97. 98. Likert Scale Assesses differences of perceptions and allows measurement and comparison of these differences Indicate your degree of agreement with this statement: “ I am an adventurous person.” (circle the best response for you)
  98. 99. Graphical Rating Scale How would you rate your enjoyment of the movie you just saw? Mark with a cross (X) not enjoyable very enjoyable
  99. 100. Semantic Differential What is your view of smoking? Tick to show your opinion. Bad ___:___:___:___:___:___:___ Good Strong ___:___:___:___:___:___:___ Weak Masculine ___:___:___:___:___:___:___ Feminine Unattractive ___:___:___:___:___:___:___ Attractive Passive ___:___:___:___:___:___:___ Active
  100. 101. Non-verbal (Idiographic) Scale Point to the face that shows how you feel about what happened to the toy.
  101. 102. Sensitivity & Reliability <ul><li>Scale should be sensitive yet reliable.
  102. 103. Watch out for too few or too many options </li></ul>
  103. 104. General aim: Maximise sensitivity (i.e. more options) Maximise reliability (i.e. less options) How many measurement options? <ul><li>Minimum = 2
  104. 105. Average = 3 to 7
  105. 106. Maximum = 10? </li></ul>Scale of Measurement Guidelines
  106. 107. FEELING ABOUT SOMETHING EXTREMELY POSITIVE EXTREMELY NEGATIVE 2-Categories GOOD NOT GOOD 3-Categories GOOD FAIR POOR 4-Categories VERY GOOD GOOD FAIR POOR 5-Categories EXCELLENT VERY GOOD GOOD FAIR POOR
  107. 108. Watch out for too many or too few responses “ Capital punishment should be reintroduced for serious crimes” 1 = Agree 2 = Disagree 1 = Very, Very Strongly Agree 7 = Slightly Disagree 2 = Very Strongly Agree 8 = Disagree 3 = Strongly Agree 9 = Strongly Disagree 4 = Agree 10 = V. Strongly Disagree 5 = Slightly Agree 11 = V, V Strongly Disagree 6 = Neutral
  108. 109. Wording Questions <ul><li>Does the question focus directly on the issue or topic to be measured? (If not, rewrite.)
  109. 110. Is the question stated as briefly as it can be? (If too long, restate it more briefly.) </li></ul>
  110. 111. Wording Questions <ul><li>Is the question expressed as clearly and simply as it can be? I (f the meaning won’t be clear to every respondent, restructure the question.)
  111. 112. Use only core vocabulary - words and phrases people use in casual speech </li></ul>
  112. 113. Wording Questions <ul><li>Limit the vocabulary so the least sophisticated respondent would be familiar with the words
  113. 114. Use simple sentences where possible and complex sentences only when actually required
  114. 115. Use two or more short, simple sentences rather than one compound or complex sentences </li></ul>
  115. 116. Finalise Questionnaire Draft <ul><li>Questions need to be exhaustive and mutually exclusive </li><ul><li>Include ‘other (please specify)’
  116. 117. Ensure categories do not overlap </li></ul></ul>
  117. 118. Finalise Questionnaire Draft <ul><li>Length </li><ul><li>Try to keep them as short as possible
  118. 119. Only ask questions that relate to objectives
  119. 120. Tricks? Font size/double sided photocopying/numbering sections </li></ul></ul>
  120. 121. Pre-testing and Pilot testing <ul><li>Pre-test – try out on convenient others & revise
  121. 122. Pilot test – try out on a small sample from the target population & revise
  122. 123. Be assertive and interactive about seeking feedback – ask questions & observe
  123. 124. “The customer is always right.” </li></ul>
  124. 125. Maximising Response Rate <ul><li>Layout and design is key
  125. 126. Respondent’s level of interest
  126. 127. Colour of paper
  127. 128. Accompanying letter / introduction
  128. 129. Mail surveys - self-addressed stamped return envelope
  129. 130. Rewards
  130. 131. Reminders or follow up calls </li></ul>
  131. 132. Examples <ul><li>Examine the examples
  132. 133. What is wrong with the questions, if anything? </li></ul>
  133. 134. Examples How old are you? ___ 18-20 ___ 20-22 ___ 22-30 ___ 30 and over
  134. 135. Examples Are you satisfied with your marriage and your job? __________________________
  135. 136. Examples You didn’t think the food was very good, did you? _____ Yes _____ No
  136. 137. Examples Environmental issues have become increasingly important in choosing hotels. Are environmental considerations an important factor when deciding on your choice of hotel accommodation? ____ Yes ____ No
  137. 138. Examples What information sources did you use to locate your restaurant for today’s meal? (please tick appropriate spaces) ____ yellow pages ____ Internet ____ word of mouth
  138. 139. Examples <ul><li>Examine questionnaire examples
  139. 140. Examine structure, design issues and question styles
  140. 141. Note cover page and details provided </li></ul>
  141. 142. Pre-test & Revise <ul><li>Pre-test items and ask for feedback
  142. 143. Revise: </li></ul><ul><ul><li>items which don’t apply to everybody
  143. 144. redundancy
  144. 145. skewed response items
  145. 146. misinterpreted items
  146. 147. non-completed items </li></ul></ul><ul><li>Reconsider ordering & layout </li></ul>
  147. 148. Survey Format Checklist <ul><li>Introduction/covering letter or verbal introducation </li><ul><li>e.g. Who are you? Are you bona fide? Purpose of survey? Ethical approval? How results will be used? Confidentiality? Further info? Complaints? </li></ul><li>Instructions </li><ul><li>Sets the “mind frame”, but be aware few people will read it without good prompting and being easy-to-read </li></ul><li>Group like questions together
  148. 149. Consider order effects, habituation, fatigue, switching between response formats </li></ul>
  149. 150. Survey Format <ul><li>Font type / size, number of pages, margins, double vs. single-siding, colour, etc.
  150. 151. Demographics - single section, usually at beginning or end of questionnaire, only use relevant questions
  151. 152. Space for comments?
  152. 153. Ending the questionnaire – say thanks!
  153. 154. Pre-test the questionnaire & revise/refine </li></ul>
  154. 155. LUNCH BREAK
  155. 156. Survey Design Critique In pairs, look through the example questionnaires, and highlight aspects which: <ul><li>could be improved
  156. 157. are particularly good
  157. 158. you would like to ask about </li></ul>
  158. 159. Implementing Surveys
  159. 160. Research Proposal <ul><li>What are your study parameters?
  160. 161. When will you carry out the research?
  161. 162. How long will data collection last?
  162. 163. Who is your target population?
  163. 164. What are your time and cost constraints?
  164. 165. What research method will be used? </li></ul>
  165. 166. Types of Questionnaires Self - administered Interview - administered Postal questionnaire Delivery and collection questionnaire Telephone survey Face to face structured interview Web-based
  166. 167. Comparison of Data Collection Methods Alreck and Settle (1995:32)
  167. 168. Sampling
  168. 169. Sampling Terminology <ul><li>Sampling Terminology
  169. 170. What is Sampling?
  170. 171. Sampling Techniques
  171. 172. Example: Shere Hite’s Sex Survey
  172. 173. Summary of Sampling Strategy </li></ul>
  173. 174. Sampling Terminology <ul><li>Population
  174. 175. Sampling Frame
  175. 176. Sample
  176. 177. Representativeness </li></ul>
  177. 178. What is a sample? <ul><li>A proportion of the research or survey population
  178. 179. Sample of a population being all ‘known cases’ </li></ul>
  179. 180. Why sample? <ul><li>Why sampling rather than a census?
  180. 181. Sampling reduces: </li><ul><li>Cost, time, sample size and defines the research
  181. 182. If the sample is representative, allows inferences to be drawn concerning the total population </li></ul></ul>
  182. 183. What is sampling? “ Sampling is the process of selecting units (e.g., people, organizations) from a population of interest so that by studying the sample we may fairly generalize our results back to the population from which they were chosen.” - Trochim, 2002
  183. 184. <ul><li>To generate a sample you need a sampling frame to select your sample from
  184. 185. List or database of all ‘known’ cases in a research population or study area (e.g., Canberra residents, Students at the University of Canberra, those staying in 5 star hotels in Sydney) </li></ul>Sampling frame
  185. 186. Sampling frame <ul><li>The set of participants from which the sample is drawn
  186. 187. Examples: </li><ul><ul><li>Electoral Rolls
  187. 188. Membership Lists (organisations, graduates association)
  188. 189. Telephone Book
  189. 190. Members of Specific Groups or Clubs (Fishing, Ramblers)
  190. 191. Households or post codes </li></ul></ul></ul>
  191. 192. Sampling frame <ul><li>What is the scope of your study? e.g., </li><ul><li>Local residents of Canberra over 18 or Canberra Households?
  192. 193. 5 Star hotel guests in last 5 years, 12 months, 2 weeks? </li></ul></ul>
  193. 194. Sampling frame <ul><li>Difficulties in identifying research population or finding a reasonable sampling frame e.g., ecstasy users
  194. 195. How accurate/up to date are lists or sample frames? Up to date? Are they missing members of the research population? e.g., Ph books
  195. 196. These issues impact on sampling and data collection techniques </li></ul>
  196. 197. Representativeness of sample depends on: <ul><li>Adequacy of sampling frame
  197. 198. Selection strategy
  198. 199. Adequacy of sample size
  199. 200. Response rate – both the % & representativeness of people in sample who actually complete survey
  200. 201. Note: It is better to have a small, good sample than a large, poor sample. </li></ul>
  201. 202. Sampling Example: Shere Hite ‘American Sexology’
  202. 203. Male-Female Relations <ul><li>Shere Hite ‘doyenne of sex polls’
  203. 204. Media furors & worldwide attention
  204. 205. 127-item questionnaire about marriage & relations between sexes
  205. 206. 4500 USA women, 14 to 85 years
  206. 207. Society and men need to change to improve lives of women </li></ul>
  207. 208. Some of Hite’s findings.... <ul><li>70% married for 5 years having affairs... </li></ul>(usually more for ‘emotional closeness’ than sex) <ul><li>76% did not feel guilty
  208. 209. 87% had a closer female friend than husband
  209. 210. 98% wanted “basic changes” to love relationships
  210. 211. only 13% married for 2+years were still in love
  211. 212. 84% were emotionally unsatisfied
  212. 213. 95% reported emotional & psychological harassment from their men </li></ul>
  213. 214. Some of the critical comments.... <ul><li>“She goes in with prejudice & comes out with a statistic.”
  214. 215. “The survey often seems merely to provide an occasion for the author’s own male-bashing diatribes.”
  215. 216. “Hite uses statistics to bolster her opinion that American women are justifiably fed up with American men.” </li></ul>
  216. 217. Response rate & Selection bias - 1 100,000 questionnaires Sent to a variety of women’s groups - feminist organisations, church groups, garden clubs, etc. 4,500 replied (4.5% return rate)
  217. 218. “ We get pretty nervous if respondents in our survey go under 70%. Respondents to surveys differ from nonrespondents in one important way: they go to the trouble of filling out what in this case was a very long, complicated, and personal questionnaire.” - Regina Herzog, University of Michigan Institute for Social Research Response rate & Selection bias - 2
  218. 219. Sampling Techniques <ul><li>Probability (random) sampling </li><ul><li>Random
  219. 220. Systematic
  220. 221. Cluster </li><ul><li>Multi-Stage Cluster </li></ul></ul><li>Non-probability sampling </li><ul><li>Quota
  221. 222. Convenience
  222. 223. Snowball </li></ul></ul>
  223. 224. Random/probability sampling <ul><li>Each unit has an equal (and perhaps known) chance or probability of selection in the sample
  224. 225. Selection occurs entirely by random chance
  225. 226. Often called representative sampling </li></ul>
  226. 227. <ul>Simple random sampling </ul><ul><li>Everyone in the target population has an equal chance of selection
  227. 228. Useful if clear study area or population is identified
  228. 229. Similar to a lottery
  229. 230. List of names are assigned #s and randomly select #s of respondents
  230. 231. Randomly select # through table of random #s or by computer </li></ul>
  231. 232. <ul>Systematic random sampling </ul><ul><li>Selecting without first numbering
  232. 233. Respondents (units) selected from a list/file.
  233. 234. Useful when survey population is similar e.g. List of Students, List of Package Tourists
  234. 235. Select sample at regular intervals from the population e.g., every 5 th person on a list </li></ul>
  235. 236. <ul>Systematic random sampling </ul><ul><li>Cannot do 1 in every 5
  236. 237. As then 4 people out of 5 stand no chance of being selected
  237. 238. Select a random starting point between 1 and 5 </li></ul>
  238. 239. <ul>Stratified Random Sampling </ul><ul><li>Sub-divide population into strata (e.g., by gender, age, or location)
  239. 240. Then random selection from within each stratum
  240. 241. Improves representativeness
  241. 242. e.g., Telephone interviews using post-code strata </li></ul>
  242. 243. <ul>Non-Random/Non-Probability </ul><ul><li>Also called purposive or judgemental sampling
  243. 244. Useful for exploratory research and case study research
  244. 245. Able to get large sample size quickly and useful when can’t find a sample frame </li></ul>
  245. 246. <ul>Non-Random/Non-Probability </ul><ul><li>Make assumptions and maybe generalisations from your data, but not on statistical grounds
  246. 247. Limitations include potential bias and applicability </li></ul>
  247. 248. <ul>Convenience Sampling </ul><ul><li>Sampling is by convenience rather than randomly
  248. 249. Due to time/financial constraints
  249. 250. e.g. surveying all those at a tourist attraction over one weekend </li></ul>
  250. 251. <ul>Purposive Sampling </ul><ul><li>Respondents selected for a particular purpose e.g., because they may be “typical” respondents
  251. 252. e.g., select sample of tourists aged 40-60 as this is the typical age group of visitors to Canberra
  252. 253. e.g., Frequent flyers to contact regarding service quality in an airline setting </li></ul>
  253. 254. Summary of sampling strategy <ul><li>Identify target population and sampling frame
  254. 255. Selection sampling method
  255. 256. Calculate required sample size
  256. 257. Maximise return rate </li></ul>
  257. 258. <ul>Task </ul>A research project's aim is – “To identify the behaviour and attitudes of UC students with regard to its computing services”. <ul><li>What is the research population?
  258. 259. How might you get hold of a sample frame?
  259. 260. What sampling technique would you use? </li></ul>
  260. 261. <ul>Confidence Levels / Margins of Error </ul><ul><li>Relates to representativeness of a sample of a target population - to what extent can we be confident about the results?
  261. 262. Gives the estimated range of values into which we expect other samples to fall say 95% of the time
  262. 263. Social sciences = at least 90% or above, preferably 95% </li></ul>
  263. 264. <ul>Example 95% Confidence Level Graphs </ul>
  264. 265. <ul>Example 95% Confidence Level Table </ul>
  265. 266. <ul>Confidence Level Example </ul>Survey of visitors to Canberra between June 1 - August 31, 2004 = 10,000 visitors Want to work with 95% confidence level and 3% margin of error How many to survey? Answer: Sample size = 964 people Q: What are your favourite attractions?
  266. 267. <ul>Following Results </ul>+ / - 3% error @ 95% confidence level
  267. 268. <ul>Confidence Intervals / Margins of Error </ul><ul><li>One time out of 20, we expect the answers may be greater than +/- 3% </li><ul><li>Establish the confidence level, margin of error you want to work with
  268. 269. Identify the number of surveys to be done
  269. 270. Once you have completed that number, do not do any additional surveys </li></ul></ul>HOWEVER….
  270. 271. <ul>Confidence Intervals / Margins of Error </ul><ul><li>Sometimes we do surveys and we do not know how many will be returned until later, as with postal surveys
  271. 272. Thus you have to calculate the margin of error afterwards…. </li><ul><li>Count up # of returned surveys
  272. 273. Identify target population and confidence level
  273. 274. Calculate margin of error </li></ul></ul>
  274. 275. <ul>Confidence Intervals / Margins of Error </ul><ul><li>All this assumes you know your target population and can get a sample frame </li><ul><li>If not, a non-random sampling technique is best </li></ul><li>Also consider whether you want to report results for sub-groups – the margins of error will be wider </li><ul><li>Consider stratified random sampling </li></ul></ul>
  275. 276. Evaluation <ul><li>Please complete the workshop evaluations </li></ul>
  276. 277. References <ul><li>Alreck, P. L. & Settle, R. B. (2003). The survey research handbook (3rd ed.). Chicago: McGraw-Hill/Irwin.
  277. 278. See also Readings </li></ul>
  278. 279. Acknowledgements This is an adaptation and extension of presention slides originally developed by (and with permission from) Dr Brent Ritchie , currently at the School of Tourism at The University of Queensland.
  279. 280. Open Office Impress <ul><li>This presentation was made using Open Office Impress.
  280. 281. Free and open source software.
  281. 282. http://www.openoffice.org/product/impress.html </li></ul>

×