Likert Scales - Reverse Scored Items (Part 2)
HTML-код
- Опубликовано: 21 авг 2024
- This video covers how to reverse code Likert items in SPSS. All steps in SPSS to successfully reverse code an item is illustrated in this 3-part video series.
Video Transcript: and if I have someone who answers with a 1, and that indicates they're highly perfectionistic, as it does on these last two questions, then what if I flipped or literally reversed the scale, so that someone who answers down here instead of a 1 they now get a 7, and someone who answers with a 2, instead of a 2 they get the reverse score, or a 6. And I show you down here how that works. So on these two items, "I never aim for perfection on my work," if I'm highly perfectionistic, then I'm going to answer a 1 once again. But, it's a perfectionism scale, so people who are high on it should have high scores. So this 1 is not going to be a 1 anymore, but it's reverse coded into a 7. So you can see that here. So a 1 becomes a 7, and as you'll see the scale is just literally reversed, or flip flopped. A 2 becomes a 6, a 3 becomes a 5, a 4 stays a 4, a 5 becomes a 3, a 6 is a 2, and a 7 is 1. OK so in SPSS these four variables correspond to the four items we just looked at. And notice that the first person answered with 6s on the first two questions, which I call Perfect1 and Perfect2, and then they answered with a 1 and 2 on Perfect3 and 4, respectively; those last two questions on the scale. So notice here this person is answering high on questions 1 and 2 and low on questions 3 and 4. If you recall that scale from earlier, that indicates that they tend to respond in a perfectionistic way. The next person answers in the most extreme way, notice that they have 7s on one and two and they have 1s on three and four. And then if we scan through here we can see some people who tend not to be perfectionistic at all, or who are very low, in that regard. And that would be this person here, person number three. They answer with a 1 and 2 on the first two questions and then a 7 and 6 on the last two. So if you read through these once again and looked at these values, that should indicate someone who is not highly perfectionistic. Now in SPSS we're going to take our two negative items here, three and four, or Perfect3 and 4, and reverse code them, where you saw that before, a 1 becomes a 7, and so on. So to do that we want to go to Transform and then we want to go to Recode into Different Variables. So go ahead and select Recode into Different Variables. And then here the Recode into Different Variables dialog box opens. Now recall we're working with Perfect3 and 4. So I'm going to go ahead and move Perfect3 over and I can move Perfect4 over as well now if I want to. Let's go and select Perfect3 and then in this Output Variable box which you see over here, I'm going to go ahead and enter a new name I'm going to call that Perfect3 and then what would I like to do is I like to put _r, and that tells me, the r tells me that it's a reverse coded item and I just put the underscore there so that stands out a little more, it's easier for me to see. So let's go ahead and click Change there. And then once I click Change, notice that it puts it over here. So this is saying Perfect3 will become Perfect3_r. And, in particular, we'll be making a new variable for that. Let's go and do Perfect4 now. So click on that and then go to the Name box and just enter Perfect4_r. Click Change. And then notice those two are there now. Now that didn't do anything in terms of SPSS to reverse code these values, but what we did was we created new variable names to hold the reverse coded values. So we're going to have a new variable, Perfect3_r, and a second variable, Perfect4_r, which if this works correctly, will show right up here. OK so next let's go ahead and reverse code those values. So to do that we go to the Old and New Values button. Go ahead and click on that.
RUclips Channel (Quantitative Specialists): / statisticsinstructor
Subscribe today!
Save yourself time by jumping to 3:47
I was seriously about to ask you about the reverse ordering, since I had difficulties with the downloaded data from qualtrics, this is very eloquent and articulated. Thanks!
+Salem Al-Sanousi
Thanks for the feedback, Salem! Please spread the word!
For users who already know what Likert scale is and are just looking for how to do the coding on SPSS you should give them a heads up that the part 1 is all old news for them!
It's easy to reverse the items, but to understand why and when to reverse, that's the key.
Thank you for explaining everything with so much detail!
get on already!!!!!
Great job, man. Thanks, it help me too much.
Thank you :)
thank you for this !!
Suppose, I administrated a questionnaire where the Likert scale started with 1 = strongly agree and ended with 5 strongly disagree to a group of 100 participants. Can you apply the same reverse procedure to the likert scale I mean to start with 5 = strongly agree to 1 = strongly disagree in all the items in a completed questionnaire without changing the participants' responses. The only thing that would change will be the order from strongly disagree = 1 to strongly agree = 5, but keep the participants' original responses. Would this effect the data results or not? Does it make any difference? Any suggestions..
Can I know does every items should have an equal amount of scale if i want to run cronbach alpha ?
what happens when the scale has -1,-2 etc as answers? Many thanks for your videos. very helpful
+marion mcginn
If it’s -2, -1, 0, 1, 2, then you would reverse these for
negative items: 2 becomes -2 and so on.
Thank you
I want to know after changing the reverse item by recoding, for ordinal data (Likert scale survey data) is it necessary to change the value in variable view?
Tayeenul Hoque
Not really. For those who want to keep good records and will be accessing the data file later, I recommend it. But if you don't care if the Likert scales have labels, it is fine not to.
Another way to think about it is that creating labels is really for us, not SPSS. While they can be quite helpful for keeping good records and interpreting output, they aren't absolutely necessary.
QS
Pleaaaaaaaaase be to the point
a wast of time ... redundancy is killing me.