Upload
benkaw
View
216
Download
0
Embed Size (px)
Citation preview
Chapter 6Using Questionnaires
Systems Analysis and DesignKendall and Kendall
Fifth Edition
Kendall & Kendall Copyright © 2002 by Prentice Hall, Inc. 6-2
Major TopicsQuestion typesScalesValidity and reliabilityFormatting the questionnaireAdministering the questionnaireWeb questionnaires
Kendall & Kendall Copyright © 2002 by Prentice Hall, Inc. 6-3
QuestionnairesQuestionnaires are useful in
gathering information from key organization members aboutAttitudesBeliefsBehaviorsCharacteristics
Kendall & Kendall Copyright © 2002 by Prentice Hall, Inc. 6-4
When to Use QuestionnairesQuestionnaires are valuable if
Organization members are widely dispersed
Many members are involved with the project
Exploratory work is neededProblem solving prior to interviews is
necessary
Kendall & Kendall Copyright © 2002 by Prentice Hall, Inc. 6-5
Question TypesQuestions are designed as either
Open-endedTry to anticipate the response you will getWell suited for getting opinionsUseful in explanatory situations
ClosedUse when all the options may be listedWhen the options are mutually exclusive
Kendall & Kendall Copyright © 2002 by Prentice Hall, Inc. 6-6
Open-Ended and Closed Questions
Open-ended ClosedSlow Speed of completion FastHigh Exploratory nature LowHigh Breadth and depth LowEasy Ease of preparation DifficultDifficult Ease of analysis Easy
Kendall & Kendall Copyright © 2002 by Prentice Hall, Inc. 6-7
Questionnaire LanguageQuestionnaire language should be
SimpleSpecificFree of biasNot patronizingTechnically accurateAddressed to those who are knowledgeableAppropriate for the reading level of the
respondent
Kendall & Kendall Copyright © 2002 by Prentice Hall, Inc. 6-8
ScalesScales are devised to
Measure the attitudes or characteristics of respondents
Have respondents act as judges for the subject of the questionnaire
Kendall & Kendall Copyright © 2002 by Prentice Hall, Inc. 6-9
Measurement ScalesThere are four different forms of
measurement scales:NominalOrdinalIntervalRatio
Kendall & Kendall Copyright © 2002 by Prentice Hall, Inc. 6-10
Nominal ScalesNominal scales are used to classify
things into categoriesIt is the weakest form of
measurementData may be totaledWhat type of software do you use the most?
1 = Word Processor2 = Spreadsheet3 = Database4 = An Email Program
Kendall & Kendall Copyright © 2002 by Prentice Hall, Inc. 6-11
Ordinal ScalesAllow classificationOrdinal scales also imply rank
orderingThere is no difference between the
importance of the choicesThe support staff of the Technical Support Group is:1. Extremely Helpful2. Very Helpful3. Moderately Helpful4. Not Very Helpful5. Not Helpful At All
Kendall & Kendall Copyright © 2002 by Prentice Hall, Inc. 6-12
Interval ScalesAn interval scale is used when the
intervals are equalThere is no absolute zeroExamples of interval scales include
the Fahrenheit or centigrade scaleHow useful is the support given by the Technical Support Group?NOT USEFUL EXTREMELY AT ALL USEFUL 1 2 3 4 5
Kendall & Kendall Copyright © 2002 by Prentice Hall, Inc. 6-13
Ratio ScalesThe intervals between numbers
are equalRatio scales have an absolute zero
Approximately how many hours do you spend on the Internet daily?0 2 4 6 8
Kendall & Kendall Copyright © 2002 by Prentice Hall, Inc. 6-14
Guidelines for Using ScalesUse a ratio scale when intervals are
equal and there is an absolute zeroUse an interval scale when intervals are
equal but there is no absolute zeroUse an ordinal scale when the intervals
are not equal but classes can be rankedUse a nominal scale when classifying
but not ranking
Kendall & Kendall Copyright © 2002 by Prentice Hall, Inc. 6-15
Validity and Reliability Questionnaires must be valid and
reliableReliability of scales refers to consistency in
responseGetting the same results if the same
questionnaire was administered again under the same conditions
Validity is the degree to which the question measures what the analyst intends to measure
Kendall & Kendall Copyright © 2002 by Prentice Hall, Inc. 6-16
Problems With Scales There are three problems
associated with poorly constructed scales:LeniencyCentral tendencyHalo effect
Kendall & Kendall Copyright © 2002 by Prentice Hall, Inc. 6-17
LeniencyCaused by easy ratersA solution is to move the
“average” category to the left or right of center
Kendall & Kendall Copyright © 2002 by Prentice Hall, Inc. 6-18
Central TendencyCentral tendency occurs when
respondents rate everything as average
Improve by making the differences smaller at the two ends
Adjusting the strength of the descriptors
Creating a scale with more points
Kendall & Kendall Copyright © 2002 by Prentice Hall, Inc. 6-19
Halo EffectWhen the impression formed in
one question carries into the next question
Solution is to place one trait and several items on each page
Kendall & Kendall Copyright © 2002 by Prentice Hall, Inc. 6-20
Formatting the QuestionnaireGood response rates can be
achieved with consistent control of questionnaireFormatStyleMeaningful orderingClustering of questions
Kendall & Kendall Copyright © 2002 by Prentice Hall, Inc. 6-21
Questionnaire FormatWhen designing questionnaires
Allow ample white spaceAllow enough space for responses to be
typed for open-ended questionsAsk respondents to clearly mark their
answersUse objectives to help determine formatBe consistent in style
Kendall & Kendall Copyright © 2002 by Prentice Hall, Inc. 6-22
Order of QuestionsMost important questions go firstSimilar topics should be clustered
togetherRandomization of questions tries the
patience of respondentsControversial questions should be
positioned after less controversial questions
Kendall & Kendall Copyright © 2002 by Prentice Hall, Inc. 6-23
Web Form QuestionnairesControls (fields) used on Web forms
Single line text boxScrolling text box, used for one or more
paragraphs of textCheck box for yes-no or true-false answersRadio button for mutually exclusive yes-no
or true-false answersDrop-down menu for selection from a listSubmit or Clear buttons
Kendall & Kendall Copyright © 2002 by Prentice Hall, Inc. 6-24
Methods of Administering the QuestionnaireMethods of administering the
questionnaire includeConvening All concerned respondents
together at one timePersonally administering the questionnaireAllowing respondents to self-administer
the questionnaireMailing questionnairesAdministering over the Web or via email
Kendall & Kendall Copyright © 2002 by Prentice Hall, Inc. 6-25
Electronically Submitting QuestionnairesAdministering a questionnaire
electronically has many benefitsReduced costsCollecting and storing the results
electronically