Likert Scale and Factorial Analysis

#1
I've got to do my own project on my final psychology degree module and i just dont know what it all means, the approach, the design etc. to take. Please someone help, i'm stressed and panicking about all this.

I'm doing a survey, using statements for participants to indicate their agreement. It'll be a 4-point Likert scale: Strongly agree, agree, disagree and strongly disagree. Each participant will answer the same set of questions. My tutor says i should use a Factoral analysis. But it seems i shouldn't for a Likert scale. I'm reading things, trying to understand things, but the truth is i just don't and all the technobabble isn't helping.

Please someone just help me!
 

noetsi

Fortran must die
#5
I've got to do my own project on my final psychology degree module and i just dont know what it all means, the approach, the design etc. to take. Please someone help, i'm stressed and panicking about all this.

I'm doing a survey, using statements for participants to indicate their agreement. It'll be a 4-point Likert scale: Strongly agree, agree, disagree and strongly disagree. Each participant will answer the same set of questions. My tutor says i should use a Factoral analysis. But it seems i shouldn't for a Likert scale. I'm reading things, trying to understand things, but the truth is i just don't and all the technobabble isn't helping.

Please someone just help me!
I am not sure what you mean by factorial? Factor analysis, linear regression, t....

If you do want to do statistics technobabble is something you will have to get used to. Including different authors using different words for the same thing or the same words for different things.
 

noetsi

Fortran must die
#7
As somebody who has gone through a stats grad program I can assure you that not all people that love stats are good at math
that amazes me. Of course you can also go through stats grad programs and be lousy at math (like say me). :p
 

Karabiner

TS Contributor
#8
Yes, I agree.
But this attitude is very common.
No problem there. But the message, "Hey there, I seek free statistical advice from
people who probably appreciate statistics. So the first thing I do is to insult statistics"
seems odd to me.

With kind regards

Karabiner
 

noetsi

Fortran must die
#9
I don't know I spent much of my life in universities and see a lot of problems with universities. You can feel that statistics is taught and done poorly and still be interested in statistics. A central problem with statistics, and academics IMHO, is that many of those who teach it are interested exclusively in the highest end things and fail to consider most are not at that level. Long series of matrices and jargon based writing just loses those who are not at that level. It like the old history writers who left French expressions in english writing because all educated people should know French. Most don't know matrices and page after page of them are useless (and I have read statistical treatments that are one page after another of matrices).

The fact, maddening to me, that there is no central agreed on nomenclature is another problem. Hierarchical regression can mean totally different things depending on who is using it. If the statistical profession (like the academic profession generally) want to write for a few hundred people then they are doing it the right way.
 

katxt

Active Member
#10
We'd like to help, despite the negative comments above, but we would need to know more.
Do you have an overall plan of what you want to discover?
Are you writing the questions or are they given to you?
Are the questions arranged in sets, each set asking about aspects of a particular topic?