PSY 3307 Lecture Notes - Lecture 8: Surveymonkey, Relate, Smallpox
SchoolUniversity of Ottawa
Course CodePSY 3307
This preview shows pages 1-2. to view the full 6 pages of the document.
Specications and writing items
Plan for today
Table of specications
Specications: what type of instrument is best?
Consider objectives, time and resources available for administration and
interpretation, and instrument framework
For evaluating vocational interests: inventory
For evaluating attitudes, values: rating scale
For making diagnosis, may need tests in combination with other information.
Checklists simple, but can’t measure ne distinctions
Who will respondents be?
Knowing who your respondents will be will guide you in writing questions and scaling
responses: have to make questions and responses appropriate for the people who are
General population, students, mothers, older people etc.
Reading level must be around a grade 6 level, try to make the question clearer, use
Knowing that will help you decide how long the questionnaire will be and the type of
-Ex. Children, clear questions and rating scale for kids might use pictures ( )
-Seniors – can’t be too long, rating scale might be shorter, lettering might be
Face to face interview
Leave a questionnaire for completion
Web based questionnaires (e.g. Survey Monkey)
Phone versus in-person interview
Personal contact does work best for getting information you need
-Establish a better rapport with the other person
Does the instrument require visual stimuli?
-Required to see a picture, use blocks, draw etc.
-Need to do it in person
Is it possible that others may get involved in the interview?
-Possibility others may interfere
-Need a quiet room
Is the instrument long?
-Get people to ll out or answer longer questionaires when in person
Only pages 1-2 are available for preview. Some parts have been intentionally blurred.
-Willing to go longer with face to face
-Need visual stimuli face attraction
Is good rapport needed?
People can complete it on their own time and their own place
Can cover a wide geographical range
Eliminate human score error or bias from in person interview
Eliminates time to put in data to evaluate it
Might miss people who don’t have access to computers (low SES) or
seniors/people who aren’t computer savy or in rural/remote areas
Can’t control who’s responding as well
Honesty and can’t conrm identity
Can’t control the environment in which they are responding
- More likely to be sent out and not lost on a colored piece of paper, should have
reminders, phone call to “warm them up”
Wide geographical range (can cover some places where there is no
Less likely to screen it out
People can completely it in their own time and in their own place
Cover people who don’t have a computer or aren’t computer savvy
Response rates are really low – 10% typically (Use Dillion)
Longer time to sore and enter data
Problem with Phone Interviews:
-High percentage of people don’t home phones
-People don’t generally trust people over the phone (more honesty)
-Distracted by other people or other activities
-Getting help, don’t know what’s going on in the background
-Response rates are going down because of caller ID (legitimately being missed)
How long will instrument be?
- Type of instrument/scaling method
Try to keep it short and sweet (unless diagnosis questionnaire)
Canadian Health Survey: need great detail on population health (physical,
mental health, depression, geographic, etc.)
- Number of sub-components and amount of detail needed
- Type of respondent
Children or older people need shorter
- Desired reliability
Generally the more questions you have the more reliable the instrument
because you get more information about the person and it can better
Table of specications
You're Reading a Preview
Unlock to view full version