Urgent: Linear Regression - Survey Design - Help needed

#1
Hi everyone,

For my master thesis, I am trying to prepare a survey on key success factors for virtual worlds.

I want to do a linear regression and have a problem in designing my online survey.

My dependent variable is the success, whereas independent ones are existence of user generated content, easiness of gameplay, etc.

Nearly all my questions relating to the independent variables are measured on a 5 point lickert scale. However, I do not have any statement or question that measures the success of a virtual world. I cannot think of one and without it I cannot do my linear regression, right?

How do I have to formulate a question in order to measure my dependent variable?

I am clueless, and furthermore I su** at maths.. So, I hope you can help me, otherwise I am pretty fu*** up for my thesis.

Thank you very much for your help in advance.

Kindest regards
 

Mean Joe

TS Contributor
#2
Hi everyone,
I am trying to prepare a survey on key success factors for virtual worlds.
However, I do not have any statement or question that measures the success of a virtual world. I cannot think of one and without it I cannot do my linear regression, right?
So you're giving a survey to people, and they will answer a question measuring the success of the game? Or are you determining the success level of the game?

Some possible ways to measure success:
# of hours invested in the game (over a typical week / over the last month / since you started)
A Yes/No question: would you call this a good (ie successful) game/recommend it to a friend?
Number of people you've met from playing the game that you wouldn't have met otherwise
Level of enjoyment from playing the game
If you are determining the success level of the game, you could look at sales/download numbers

Good luck on your thesis.
 
#3
You could change your direction to be a description of games that includes how people percieve success factors in games. Here you would be giving up any sense of dependent or independent immediately. Your first technique would then be Principle Component Analysis followed by clustering analysis I think.

From what I understand when you see these surveys Principle Component Analysis is almost always the first analysis we use. From there anything might happen.
 
#4
hello and thanks for your answer.

in order to best explain you what the survey looks like, here you will find the survey:

http://www.surveymonkey.com/s.aspx?sm=I0ohF93Gt4SXGEt3dxB9qg_3d_3d

I want to determine what the key success factors are for a virtual world IN GENERAL. (this means that I cannot ask: what about this world and the other one?) more specifically I am trying to find out the key success factors for virtual environments such as Second Life, Habbo Hotel which focus more on socializing rather than gaming aspect.

My hypotheses are that user generated content, community management, etc. all those variables are key success factors. I asked questions on a 5 point lickert scale and then want to do a regression analysis in order to verify my hypotheses.

How can I do that though? Sorry, I am very bad at maths...
It would be cool if you could give me a feedback on the survey, if it is well designed.

I hope you can help me.


Thanks in advance.

Kindest regards.
 
#6
hello,

i think I got finally an idea how to measure success of a virtual world:

Time spent per connection on a virtual world
<30, 30-1, 1-2, 2-3, >3h

Frequence of connection per month
once a month, once a week, twice a week, four times a week, every day

How much have you spent on virtual worlds in the last month?
0, 1-3, 3-5, 5-10, >10

Those are all quantitative questions right?

However, I wanted to add another qualitative question to it:

you alrerady recommended a virtual world to one of your friends, family,etc.
Strongly disagree, disagree, neither disagree nor agree, agree, strongly agree

Now my question is: Do my questions hqve to be all qualitative or all quantitative? Are my questions good like this and do they measure success?

Thanks for your help.