You are Here: Home

# iv, dv ? anova, t test?? watch

1. Im a 2nd yr student ripping my hair out... 2 Iv's - gender, Male-female and social class, lots of levels, dv, fear of victimization, likert style (british crime survey) what test to run? i have looked at every text book an have no comparison to use...
2. Factorial ANOVA - 2 x how ever many levels of social class.

Your DV sucks somewhat as a continuous variable, but if you check assumptions and make an offering to the FSM you'll probably be fine.
3. Thanks, we have been given a naff data set and have to come up with a proposal and write it up.
4. I have now been told to keep the 2, iv's and combine all the dv's into 1 dv. how do i do that?
5. Your DV is fear of victimisation. That's one DV.

Your two IVs are Gender, with two levels, and Social Class, with many levels.

So you'll be wanting to do a within-subjects ANOVA. (As subjects will be in at least two of your levels.)
6. Thanks for that but how do I combine all the dv's into 1, as there were 8 questions relating to diff accounts of fear eg, walking alone, fear being attacked etc
7. Are you doing this on SPSS?

PS you're terming is wrong, be careful with that. You have 1 DV, fear of victimisation.

Those at questions are not your DV, they are the items that add up for you DV.
8. (Original post by Millie-mollie)
Thanks for that but how do I combine all the dv's into 1, as there were 8 questions relating to diff accounts of fear eg, walking alone, fear being attacked etc
Try to see this as comparable to measure of Depression - lots of questions that effectively tap one underlying dimension/factor. For instance, the BDI measures various aspects of depression-related behaviours and individual items are summed to give an overall 'depression' (BDI) score.

In your case, the dimension/factor is Fear of Victimisation, and will act as your DV. So sum up each of the individual items and derive a total score. This will represent the factor FoV. You might also want to check the reliability of the scale (use Cronbach alpha). This will tell you how reliability the FoV measure is, and whether some items are crap (and should be removed).

In SPSS, just use Compute variable, name a new variable (FoV), and tell SPSS to sum up all the individual items (i.e., Item1 + item2 + item3...+item x). The compute variable option just acts like a calculator and will give you the new variable you specify.
9. yes its on spss, so how do I add them all up to be my dv
10. (Original post by Millie-mollie)
yes its on spss, so how do I add them all up to be my dv
If you have variable columns in SPSS that represent scores on individual items, then use compute variable. You need to follow this path in the options:

Transform > Compute Variable

The dialog box will ask you to name a new variable (label it something that represents Fear of Victimisation - FoV would do). In the numeric expression section (which is just calculator-like) define the nature of the new variable - you tell SPSS to add up all the individual items. Then click 'OK', and you'll see a new variable in the datafile - FoV.

FoV is the total score across the eight items and will represent Fear of Victimisation. Your DV.
11. Thank you, thank you, I thought I would need to add them all up individually by hand...... X
12. (Original post by aliluvschoc)
Your DV is fear of victimisation. That's one DV.

Your two IVs are Gender, with two levels, and Social Class, with many levels.

So you'll be wanting to do a within-subjects ANOVA. (As subjects will be in at least two of your levels.)
If you're wanting to use a within-subjects design using gender as an IV, you must be recruiting some very dedicated subjects if they're willing to go through the op.
13. (Original post by GodspeedGehenna)
If you're wanting to use a within-subjects design using gender as an IV, you must be recruiting some very dedicated subjects if they're willing to go through the op.
No I don't think so.

I assumed it would be within-subject (not entirely within-subjects as the participants are not in all of the levels, but nevertheless...), as the participants are involved in more than one IV. If they are a female with a low social class, they are in both the Gender IV, and the Social IV.

It cannot be completely between-subjects surely? As in that case, there would be a completely new set of participants in each level of each IV?

It may be wrong but this is how we were told to set up such an experiment in the statistics/SPSS sessions at uni.

Well I'm all confused now. I might have just failed my SPSS exam.
14. (Original post by aliluvschoc)
No I don't think so.

I assumed it would be within-subject (not entirely within-subjects as the participants are not in all of the levels, but nevertheless...), as the participants are involved in more than one IV. If they are a female with a low social class, they are in both the Gender IV, and the Social IV.

It cannot be completely between-subjects surely? As in that case, there would be a completely new set of participants in each level of each IV?

It may be wrong but this is how we were told to set up such an experiment in the statistics/SPSS sessions at uni.

Well I'm all confused now. I might have just failed my SPSS exam.
Well yeah, there are completely new participants for each level of the IV. There has to be. You can't be male and female. Similarly, you can't be lower class and middle class.

In these instances, you would have four groups (bit more complex for OPs original example, but for demonstration purposes):
Lower Class Male
Lower Class Female
Middle Class Male
Middle Class Female

They're all seperate samples, so both IVs are independent (or between-subjects). Although participants have experienced more than one IV, they have only been in one condition associated with each IV.
15. (Original post by aliluvschoc)
It may be wrong but this is how we were told to set up such an experiment in the statistics/SPSS sessions at uni.
If so, they gave you duff information.

Firstly, you would determine within/related/repeated and between/unrelated/independent by the nature of the levels of each of the independent variables (not over them).

That is, each IV would be classed as independent groups or repeated-measures. Then taking the whole design into account it can be classed as independent, RM, or mixed-design depending on the nature of the IVs.

e.g.,

2 independent IVs = two-way independent ANOVA
2 repeated-measures IVs = two-way repeated-measures ANOVA
1 independent IV, 1 repeated-measures IV = two-way mixed ANOVA

Well I'm all confused now. I might have just failed my SPSS exam.
Stuff happens.
16. (Original post by GodspeedGehenna)
x

(Original post by Psych!)
Stuff happens.
fair enough guys, I guess I need to look over those notes again.

### Related university courses

TSR Support Team

We have a brilliant team of more than 60 Support Team members looking after discussions on The Student Room, helping to make it a fun, safe and useful place to hang out.

This forum is supported by:
Updated: April 5, 2011
The home of Results and Clearing

### 1,679

people online now

### 1,567,000

students helped last year
Today on TSR

### University open days

1. Sheffield Hallam University
Tue, 21 Aug '18
2. Bournemouth University
Wed, 22 Aug '18
3. University of Buckingham
Thu, 23 Aug '18
Poll
Useful resources

Can you help? Study Help unanswered threadsStudy Help rules and posting guidelines

## Groups associated with this forum:

View associated groups

The Student Room, Get Revising and Marked by Teachers are trading names of The Student Room Group Ltd.

Register Number: 04666380 (England and Wales), VAT No. 806 8067 22 Registered Office: International House, Queens Road, Brighton, BN1 3XE