[meteor_slideshow slideshow=”adssa” metadata=”height: 126, width: 630″]
This chapter has 3 parts;
1- Drafting the survey questions to ask Saudi & Foreign Drivers in KSA,
• I need 40-50 questions .
• ”I need the questions in 5 days so I can send them to my supervisors to comment on them”
Types of questions
The scale questions ( bi-polar scale) are a way of introducing high-powered statistical techniques. You do need some yes /no questions; you do need some multiple choice questions because they become predictors.
2. Yes /No
3. Multiple Choice
With gender it is either male or female. You can use those two categories for their tests – for the scalar questions, the attitudinal questions, and the opinions of other people. You use the demographics to separate people out into two different groups. You’ve got all the women who are entering these scales and you have all the men who are entering these scales. And you can find out if there is a major difference on those items that are scaled. But you can’t do that with any of the questions that are multiple choice. You can only talk about 20% or 15% etc. You can compare them, but you can’t subject them to any tests.
Now the tests. I suggest you try as much as possible to use scales. Not any old scales. It is significant how you construct these scales. For example, you could say “this particular advert was a really effective example of marketing or social awareness”. Then you put down a scale –strongly agree, agree, no opinion, disagree, strongly disagree.
You actually bias the way that people answer the question by the way you phrase your original statement. So people looking for effectiveness of campaigns, if you put up a negative example- such as “this was not viewed in a particular way” then you bias it towards the questioner. So the actual nature of the question and how you phrase the question, is important. So typically what happens with badly worded questions, or biased questions, you start to bias your survey. And people do it all the time in terms of marketing and advertising. When they ask questions of people, they are not fair questions. They know that they are going to get certain kinds of answers, over a large number of people. And that comes out mathematically. Most of the punters out there, including govt departments are not aware of this. The form of the questions you ask will have a dramatic affect on the answers you will get….. I always recommend to my students a bi-polar scale. You have a negative statement at one end, and a positive statement at the other end. The problem with this is that it takes time, and if your participants are educated and take the time over the question. It’s harder for them to make a judgement – they have to read both sides of the scale. It gets repetitive. There is a whole page of positive, negative, positive, negative…. It is really boring to scan. In the end people may give up, or just go tick tick tick. When you do a page properly, you must read the left and the right side of the page (one has negative statement ,the other positive). I put all the positives on the right (7’s), all the negatives on the left (ones). In Arabic it is different – you do it the other way around. This is the fairest way to do it.
• Scale – strongly agree, agree, no opinion, disagree, strongly disagree.
• Use bi-polar scale ( ie phrase statement positively and then negatively
Mix the questions- break them up
You can also break up the survey in how it appears visually. But if you intersperse them, that is the smartest way, because people won’t get that kind of fatigue.
Why 40 questions?
The more questions you ask the better. You can throw some away. Just because you ask 40 questions , doesn’t mean you have to use them all. You are going to throw 20% away. You don’t know which ones are going to be useful. 100 is too many. 40 or 50 is right. (10-20 is too small). 20-40 is good.
For good ideas – look online
Look online for survey questions about road safety campaigns or awareness campaigns in general because there is no copyright for questions.
Please check http://stc.uws.edu.au/CR12
or if that is down http://stc2.uws.edu.au/CR
It has all information about survey and questions my supervisor wants me to follow.
2- After collecting the data, enter the data in SPSS data file.
3- Analyse the qualitative data using SPSS software.
[meteor_slideshow slideshow=”best” metadata=”height: 126, width: 630″]