I'd like to bring up a few issues related to the phrasing of items and scale measurement here. I guess it will be useful to start thinking about them now.
SCALE PHRASING
1. Barriers
Do you think we need to include barriers at all in our phrasing?
I'm uncertain about this because there are many situation in group work related to positive aspects rather then negative (Michael's item is a good example). And I don't think we can really phrase all of them using barriers. My suggestion would be to come up with the items first and see if it makes sense to insert barriers in them.
2. Reverse phrasing
Do we need to include this and why?
My understanding is that they are used to check the accuracy and attenrion of respondents. But do we really need to do this?
SCALE MEASUREMENT
- 1-5 scale
- 1-7 scale
- 1-10 scale
- 0 - 100 scale
Which one is better and why?
I'm against 0-100 because I don't believe that 0 is meaningful in this context (we all have self-efficacy, but it can be very-very low) and, as Yeoeun suggested, it might be unreliable in statistical analysis.
I think 1-10 is too differentiated but is better than 0-100 because it doesn't include 0.
1-5 scale is okay for me, but I'm more for 1-7 scale because I believe that it's big enough to differenciate subtle aspects of variation in respondents' attitudes (better than 1-5), but not too cumbersome. There's also evidence from this article.
- "Reliability is lower for scales with only two or three points compared to those with more points, but suggest that the gain in reliability levels off after about 7 points"
- "Validity is higher for scales with a moderate number of points than for scales with fewer, with the suggestion that validity is compromised by especially long scales".
What're your thoughts on this, guys?
Hi Ira,
ReplyDeleteGreat points. I do think we should include barriers! I agree that the challenging part of identifying barriers is coming up with barriers that are general enough to apply to many different groups.
I think I like the idea of a 1-7 scale. I think 1-7 might capture a little more variation than 1-5 without being as complicated as 1-10.
Anna,
DeleteDo you think we need to come up with barriers first and phrase items based on that or we should create items first and then rephrase them?
Great, I also support 1-7. I wonder what the others think about this.
Ira,
DeleteI don't know if we need to (I think your items were really good without coming up with barriers first), but I think it would make it easier to come up with lots of items.
I also like 1-7 scale the most, if this is for college students or adults. I think 1-5 scale is better if it's younger children because they may not be able to differentiate those. I still want to try 1-10, but not 1-100 because of my past failure...
ReplyDeleteAlso, I think it would be great if we can include some items related to barriers in our measure. If it is hard to include barriers in all of our items, we can have "overcoming barriers" component, as part of our measure, with items dealing with possible barriers of collaboration.
I think that's a great idea. I should've specified -- I don't think we need barriers in every item, but I do think there are obstacles that are faced in collaboration that might impede a groups ability to work together effectively.
Delete1-7 scale and reversed scoring for the reasons you mentioned would be my choice.
ReplyDelete