Download presentation
Presentation is loading. Please wait.
1
Survey Design & Item Construction Lindsay Couzens, M.S. UNLV’s 2009 Academic Assessment Workshop May 14 th & 15 th
2
What is your overall research question? What questions do you want your survey to answer? Is there an existing survey that answers your questions? Keep length of the survey in mind Define target group Identify Objectives 2 UNLV Office of Academic Assessment
3
Identify Objectives Based on purpose/research question Example: Fowler, 1995 Example: Fowler, 1995 Develop definitions for objectives All respondents should have a common understanding of the meaning of each question Specify the kinds of answers needed to meet the objectives Informs item construction 3 UNLV Office of Academic Assessment
4
In person Can clarify confusing statements Immediate feedback Limited to those who are present Mail Can reach large geographic areas Depend on respondents’ motivation Requires current address database Email Can reach large geographic areas instantly Cost-efficient Requires computer access Online Cost-efficient Requires computer access Interview Can ask exploratory or probing questions Can assist with unclear questions Requires trained interviewers Requires up-to-date telephone numbers 4 UNLV Office of Academic Assessment
5
Timeline Number of data collection points Cutoff dates Cost Who will construct the survey? How? ▪ Survey Monkey ▪ Elisten ▪ WebCampus tool 5 UNLV Office of Academic Assessment
6
How will the survey be distributed? Who will analyze the survey? Do you need special hardware or software? Who will maintain the survey? Who is responsible for updating, revising, editing? UNLV Office of Academic Assessment 6
7
Pilot testing Focus groups How will the surveys be distributed? How will they be collected? 7 UNLV Office of Academic Assessment
8
What is the research question? Do the objectives for the survey flow from the research question? Is each objective stated in concrete terms? Does each objective measure one idea or concept? UNLV Office of Academic Assessment 8
9
Item Construction
10
Question Content Each question should be mapped to one of your identified and defined objectives Consider the content, scope and purpose of each item Response formats Open-ended or closed-ended? Survey layout Items should be grouped by topic AND by response type 10 UNLV Office of Academic Assessment
11
Response Formats Things to Consider Clarity of Directions Levels of measurement ▪ Nominal ▪ Ordinal ▪ Interval Exhaustiveness ▪ Response alternatives have sufficient range 11 UNLV Office of Academic Assessment
12
Response Formats Things to Consider Exclusiveness ▪ Response alternatives are mutually exclusive Balancing Categories ▪ There should be an even number of alternatives on both sides of neutral UNLV Office of Academic Assessment 12
13
Response Formats Numerical Rating Scales Likert-type scales ▪ Rates attitudes or opinions Horizontal rating scales ▪ Used to indicate where a respondent’s attitude or opinion falls between two opposite attitude positions 13 UNLV Office of Academic Assessment
14
Response Formats Numerical Rating Scales Semantic differential ▪ Opposite adjectives rather than attitude positions Vertical rating ladder ▪ Ranking UNLV Office of Academic Assessment 14
15
Response Formats Scores Out of 10 ▪ Respondents rate each item Ranking Items are ranked relative to other items in the set 15 UNLV Office of Academic Assessment
16
Response Formats Checklists Respondent checks all that apply Binary choice formats Dichotomous questions ▪ Ex: What is your sex? (M/F); Do you smoke? (Y/N) Paired comparisons ▪ Respondent chooses between two overlapping options UNLV Office of Academic Assessment 16
17
Response Formats Multiple choice formats Choice between multiple nominal categories ▪ Marital status Choice between ordinal categories ▪ How often do you eat in the dining commons? 17 UNLV Office of Academic Assessment
18
Response Formats Multiple choice formats Choice between ordered attitude statements ▪ Which attitude statement most closely matches your own? Numerical answers ▪ Salary; how many people live in the household UNLV Office of Academic Assessment 18
19
Question Wording Checklist deVaus, 2002 Is the language simple? Can the question be shortened? Is the question double-barreled? Is the question leading? Is the question negative? 19 UNLV Office of Academic Assessment
20
Question Wording Checklist deVaus, 2002 Is the respondent likely to have the necessary knowledge? Will the words have the same meaning for everyone? Is there a prestige bias? Is the question ambiguous? 20 UNLV Office of Academic Assessment
21
Question Wording Checklist deVaus, 2002 Is the question too precise? Is the frame of reference for the question sufficiently clear? Does the question artificially create opinions? Is personal or impersonal wording preferable? 21 UNLV Office of Academic Assessment
22
Question Wording Checklist deVaus, 2002 Is the question wording unnecessarily detailed or objectionable? Does the question have dangling alternatives? Does the question contain gratuitous qualifiers? Is the question a “dead giveaway”? 22 UNLV Office of Academic Assessment
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.