DESIGNING EVALUATION INSTRUMENTS

1
DESIGNING EVALUATION INSTRUMENTS

• Class X
• AEE 577

2

• Upon completion of this lesson, students should
  be able to
• List the step by step procedures for developing
  quality evaluation instruments
• Describe the errors that must be controlled in
  evaluation instruments
• Develop different forms of questions to record
  outcomes such as change of knowledge, attitudes,
  skills, aspirations, and behaviors
• Write process evaluation questions
• Describe reliability and validity
• Identify double barreled questions and
• Develop an evaluation instrument.

3
How to Design Your Data Collecting Instrument?

• Where to begin?

4
Begin with the information needs of key
stakeholders

• Information needs for program improvement
• Information needs for accountability

5
Designing Instruments Step 1 Identify the type
of data and information you need to collect.

• Focus on the information needs of key
  stakeholders.
• Clearly identify what data and information are
  needed to collect for this purpose.
• Identify the major categories of information that
  you need to collect.
• List subcategories of information under major
  categories.

6
Designing Instruments Step 2 Develop the
Sketch of Your Instrument.

• List the major items in your instrument to
  structure it.
• Organize the structure of your instrument to
  collect needed data.
• Organize subcategories under each major topic.
• Include the demographic data collection section
  at the very end of the instrument.

7
Designing Instruments Step 3 Identify Necessary
Scales and Questions.

• Determine the scales you need to include in your
  instrument.
• Determine the types of questions you need ask.

8
Designing Instruments Step 4 Be Consistent in
Numbering Answer Choices and Scales

• It is a good idea to use low numbers for lower
  manifestation of a measuring variable.
• For example levels of education answer choices
  should be coded as follows
• High school diploma
• Bachelors degree
• Masters degree
• Doctorate
• By using a consistent pattern throughout the
  instrument you can easily interpret results.

9
Designing Instruments Step 5 Writing Questions

• As a general rule, when writing questions, you
  must ask why am I asking this question?
• Remember your evaluation information needs
  always.
• Think about the answer before you write any
  question.
• There are two ways to write a question
• Open-ended
• Example What methods do you use to educate
  farmers on sustainable agriculture?
• Closed ended
• Example What methods do you use to educate
  farmers on sustainable agriculture?
• Field days
• Workshops
• Seminars
• Printed materials
• Electronic materials
• Others (please specify)___________________

10
Designing Instruments Writing Open-Ended
Questions

• Things to remember when writing questions
• Write questions clearly and concisely.
• Start with least sensitive or non-threatening
  questions.
• Write questions by thinking about the reading
  level of the target population.
• Avoid double negatives.
• Avoid double-barreled questions.
• Example Are you satisfied with the place and
  time of the program?

11
Designing Instruments Writing Open-Ended Questions

• Open-ended questions are useful to explore a
  topic in depth.
• However, open-ended questions are difficult to
• Analyze
• Respond
• Therefore, limit the number of open-ended
  questions to the needed minimum.
• When you need to ask a sensitive question it is
  appropriate to use a closed-ended question with
  response categories for the sensitive
  information.
• Example Asking income or age (Ask what is your
  age group and provide age categories instead of
  asking how old are you?)

12
Designing Instruments Writing Closed-Ended
Questions

• When writing closed-ended questions
• Make sure to include all possible response
  categories.
• If you have not included all possible answer
  categories, it is a good idea to include a
  category called Other and provide instruction
  to specify what the respondent means under this
  category.
• Make sure that your answer categories are
  mutually exclusive.
• Example What is your age group?
• Less than 20 years
• 20-30 years
• 31-40 years
• 41-50 years
• Above 50 years

13
Designing Instruments Writing Closed-Ended
Questions

• Closed-ended questions are
• Easy to analyze.
• Not exploratory in terms of searching
  information.

14
Scale Development

• Develop scales if you need to include in your
  instrument.

15
Guidelines For Scale Development

• Scales are developed to measure elusive phenomena
  that cannot be observed directly. Example
  Attitudes, Aspirations.
• Therefore, scale development should be based on
  the theories related to the phenomenon to be
  measured.
• Thinking clearly about the content of a scale
  requires thinking clearly about the construct
  being measured.

16
Guidelines For Scale DevelopmentGenerate an Item
Pool

• The properties of a scale are determined by the
  items that make it up.
• At this stage, you need to develop more items
  than you plan to include in the final scale.

17
Characteristics of Good Items

• Unambiguous.
• Avoid exceptionally lengthy items.
• Consider reading levels of the target
  respondents.
• Include positively and negatively worded items.
  The purpose of wording items both positively and
  negatively within the same scale is usually to
  avoid acquiescence, affirmation, or agreement
  bias.

18
Guidelines For Scale DevelopmentDetermine the
Format for Measurement

• There are different formats
• Identify the format you would like to use with
  your items.
• Determine how many response categories you need
  to include in your format.

19
Guidelines For Scale DevelopmentDetermine the
Format for Measurement

• The number of response categories should be
  limited to the respondents ability to
  discriminate meaningfully.
• Normally 5-7 response categories are adequate for
  extension and education program evaluations.
• Example 1. Strongly disagree 2. Disagree 3.
  Somewhat agree 4. Agree 5.Strongly agree

20
Guidelines For Scale Development Likert Scale

• Named after Rensis Likert.
• This is the most common format
• The response options should be worded so as to
  have roughly equal intervals with respect to
  agreement. That is to say the difference in
  agreement between any adjacent pair of responses
  should be about the same as for any other
  adjacent pair of response options.
• Common choices for a mid point include neither
  agree nor disagree and Neutral.

21
Guidelines for Scale Development Likert Scale

• Example for items in Likert format
• Strongly Disagree
• Disagree
• Somewhat Agree
• Agree
• Strongly Agree

22
Guidelines For Scale DevelopmentSemantic
Differential Scaling

• There are several numbers between the adjectives
  that constitute the response options.
• Example The quality of training session
• Poor 1 2 3 4 5 6 7 Excellent

23
Instrument DevelopmentStep 6 Provide Necessary
Instructions to Complete the Survey

• Clear instruction is essential to facilitate the
  responding process.
• Instructions should be clearly and politely
  stated.
• Clear instructions increase your return rate as
  well as accuracy of your data.

24
When You Develop a Questionnaire

• Keep it short, simple, and clear
• Include only needed questions for indicators
• Should be compatible with the reading level of
  the respondents
• When you use closed-ended questions make sure to
  include all possible answer choices.

25
Instrument DevelopmentStep 7 Format Your
Instrument

• Appearance and editing of your instrument are
  important determinants of response rate.
• Therefore, format, structure, and edit your
  instrument professionally.

26
Instrument DevelopmentStep 8 Establish Validity
and Reliability of Your Instrument

• Reliability refers to the extent to which a
  measuring instrument is consistent in measuring
  what it measures.
• Test-retest method We administer the instrument
  to a sample of subjects on two occasions and
  correlate the paired scores to establish the
  reliability.
• Validity refers to the extent to which an
  instrument measures what it is intended to
  measure.
• Use experts views to establish validity.

27
APPLICATION OF STEPS
28
Determine Your Evaluation Questions

• Identify the precise questions need to be
  answered.
• Use the logic model to narrow the focus of
  evaluation.

29
LOGIC MODEL
Measuring Program Impact
INPUTS OUTPUTS OUTCOMES -
IMPACT
Activities Participation LEARNING ACTION
IMPACT
What resources does your program need to achieve
its objectives?
What should you do in order to achieve program
goals and objectives?
Who should - participate - be
involved? - be reached?
What do you expect the participants will know,
feel or be able to do immediately after the
program?
What do you expect that participants will do
differently after the program?
What kind of impact can result if the
participants behave or act differently?
Staff Volunteer Time Money Materials Equipment
Technology Partners
Workshops Meetings Camps
Demonstrations Publications Media
Web site Projects Field Days
Number of target clients Their
characteristics Their reactions
Awareness Knowledge Attitudes Skills
Aspirations
Behavior Practice Decisions
Policies Social Action
Social Economic Environmental
30
Possible Question Categories

• Process evaluation questions (These are mostly
  open-ended)
• Questions on client characteristics
• Are we reaching the target clients?
• Questions on program delivery
• Did the program use effective teaching methods?
• What are the weaknesses?
• Impact evaluation questions
• Questions on clients satisfaction
• Did the target clients find the program useful?
• Outcomes
• Did the program participants change KASA?
• Did the program participants change their
  practices?
• Impacts
• Did the participants save money/improve health
  condition?

31
What Data Are Needed for Program Improvement?

• Were participants satisfied with
• Information received
• Instructors
• Facilities
• Quality of training
• What do they like/dislike about the training
• Did the training meet their expectations?
• If not, Why
• Ideas for further improvement
• Look for data that you can use to fix weaknesses
  and build on strengths.

32
How to Collect Training Improvement Data?
Please circle the appropriate number for your
level of response.
33
How to Collect Training Improvement Data?

• Did the training session meet your expectation?
• Yes
• No
• Would you recommend this training workshop to
  others?
• 1. Yes
• 2. No
• If not, why____________________________________
• What did you like the most about this training?
• What did you like the least about this training?
• How could this training be further improved?

34
Other Data

• Demographics
• What is your gender?
• ____ Male____ Female
• How do you identify yourself?
• ___African American
• ___American Indian/Alaskan
• ___Asian
• ___Hispanic/Latino
• ___Native Hawaiian/Pacific Islander
• ___White
• ___Other

35
What Data Are Needed for Program Accountability?

• You need impact data
• To prove that your program achieved its
  objectives

36
How to Document Perceived Knowledge Change?
Example for Agriculture
37
How to Document Levels of Aspirations?

• At the end of a successful training session,
  participants will have a heightened level of
  aspirations to apply what they learned.
• They are ready to taking charge of what they
  learned.
• Participants are asked whether they intend to
  apply what they learned.
• Example As a result of this training, do you
  intend to drink reduced fat milk? The answers to
  this question would be
• No
• Maybe
• Yes
• Im already doing this

38
How to Document Aspirations?
Example for FCS
Please circle the number that best describes your
answer.
39
Retrospective Pre and Post Evaluations

• Advantages
• Simple Easy to collect data
• Disadvantages
• Not appropriate for collecting data from very
  young audiences and low literacy adult audiences.
  Because they will not be able to compare before
  and after situation retrospectively.

40
Pre and Post Evaluations

• Pre Evaluation is administrated before your
  training session.
• Post Evaluation is administrated at the end of
  your training session.
• You need to match pre and post evaluations for
  comparison.
• Pre and Post Evaluation will document three
  impact indicators
• Change in Knowledge
• Change in Skills
• Levels of Aspirations

41
How to Document Change in Knowledge?

• Ask same set of questions before and after your
  educational session and compare their answers to
  document the knowledge gain from the program.

42
How to Document Change in Knowledge?
Example for FCS
43
How to Write Knowledge Testing Questions

• Dont use general knowledge questions.
• Dont include attitudinal or perceptual
  statements.
• Example Growers should practice conservation
  tillage. __True __False

44
True and False Questions vs Multiple Choice
Questions

• True and False questions save your time and
  respondents time.
• Easy to analyze.
• Help you keep your survey short.

45
How to Document Change in Skills?

• Skill changes are measured indirectly by using
  participants levels of confidence to carry out
  the specific tasks focused in the program.
  Example Participants confidence to calibrate
  equipment.

46
How to Document Change in Skills?

• We record their levels of confidence to carry out
  specific tasks before and after the program on a
  Likert scale.
• Compare pre and post responses to document the
  skill change.

47
How to Document Change in Skills?
Example for Agriculture
48
Pre and Post Evaluations

• Advantages
• Comparison is more accurate than retrospective
  pre and post evaluation.
• Appropriate for young and low reading audiences.
• Disadvantages
• If you want to compare pre and post evaluations
  to assess the change you must match pre and post
  evaluations for each participant.
• This is somewhat challenging.

49
Change Attitudes

• Difficult to measure
• Need to be very careful in designing scales to
  measure attitudes
• Not a practical indicator
• Pre/Post tests

50
CHECKING ATTITUDES
To what extent do you agree or disagree with each
of the following statements
Statement Strongly
Disagree Undecided Agree Strongly Disagre
e Agree 1 2
3 4 5

• Conservation Tillage is profitable. 1
  2 3 4 5
• (Need to include at least 10-15 items to achieve
  desired level of validity and reliability)

51
How to Document Behavior Change?

• You need to understand the behavior change
  process for designing evaluation questions.

52
Understanding Behavior Change Process

• Behavior change is a process.
• Prochaska and DiClemente developed a model to
  explain the human behavior change process. This
  model is called the Transtheoretical Model.
• According to the Transtheoretical Model, there
  are five stages in behavior change process.

53
Prochaska and DiClementes Stages of Change
Prochaska, J. O. and DiClemente, C. C. (1994).
The Transtheoritical Approach Crossing
Traditional Boundaries of Therapy.
Malabar, Florida Kerieger Publishing Company.
54
Evaluation Template
For each of the following practices, please
circle the number that best describes your
current behavior.
55
How to Collect Impact Data from Multi-Session
Programs

• Benchmark Survey is administrated before the
  Extension program.
• End of Program Survey is administrated at the
  end of the extension program.
• By comparing benchmark and end of program surveys
  you will be able to document the change of
  participants behaviors/practices and skills.