What statistic method using spss to use to analyze online buying behavior based on different variables? I have the data with me however I don't have any idea as to how to analyze the variables. Can anybody help me? The variables/questions are as listed:
1. Gender   
2. Age
3. Education Level
4. Occupation
5. Household monthly income (in PHP $)  
6. Daily usage of the Internet  
7. Have you ever used the internet for purchase? (If your answer is YES, please skip questions 14-20. If your answer is NO, please skip questions 8-13 and jump to the next page)   
8. How many times have you purchased online?

    "Reasons for adopting the online purchases

9. Lower prices
10. Easiness of online buying procedures
11. Wide variety of products    
12. Various payment options 
13. High quality of products    
14. Security and privacy reasons    
15. Need to physically examine the product  
16. Prefer to buy from retail stores    
17. Do not own a credit card    

"Reasons for not adopting the online purchases
18. Are unaware of the buying procedure through the Internet    
19. Shipping delays     
20. Unaffordable shipping fees  
General expectations:
21. Online stores should have a good reputation 
22. Online stores should provide adequate payment options.  
23. Delivery service should be reliable     
24. The internet consumers must be protected by government laws.    
25. Consumers can return a product and get refunded the purchase price  
26. Security mechanisms are used to ensure personal data safety 
27. The company has also retail stores  
28. Online stores offer guarantee for their provided products   
29. Shop any time of the day    
30. Shop abroad 
31. Saves time  
32. Easy to do comparison shopping between products, as well as, online stores  
33. More easily you can find a product compared to retail stores    
34. Have more options compared to retail stores     
35. Easy to find real bargains  
36. Provided products are cheaper compared to retail stores 
37. The whole buying procedure is easier compared to retail stores  
38. Consumers can find products that are not in retail stores   
39. Have much more time to evaluate and select a product    
40. Online stores promise more than they can practically offer  
41. Consumers can not completely trust online stores    
42. Online stores are not always official representatives of their offered products 
43. Consumers find it difficult to confirm the reliability of the provided products 
44. It is possible to have your credit card data intercepted    
45. It is difficult to have after-sales services

frequency of use for thee purposes: 
46. Surfing (e.g. read news/articles)   
47. Social Networking   
48. E-mail  
49. Job search  
50. Research    
51. Chatting (e.g. skype/yahoo messenger)   
52. Playing Online Games    
53. Searching for product info

 A: *

*Have an objective: With 53 variables, each of which can be reported on singly or in analyses that combine two or more variables, there is an infinite range of analysis you can do. While there is an enormous range of possible objectives, I'll invent a simplistic example. Let's say you are consultant to a hip local clothing company building its first online retail presence, and they asked you to help them understand their likely target customers and design a marketing plan to bring traffic to the website and ultimately sell something. 

*Break down your objective into components: The objective as stated as too big to connect directly to the data. You've got to break it down.


*

*Understand target customer: a hip clothing company can probably tell you the age, gender, and income range of their current customers. You can tell them what percentage of the population surveyed meets those criteria, and the other demographics associated with that target customer.

*Bring traffic to the website: you can tell them how frequently those target customers use different categories of online media (email, jobs, etc.) which will help them come up with an initial display advertising plan. 

*Sell them something: you can tell them the 5 most important and least important things to those target customers who shop online.


*Analyze the data: 


*

*Understand target customer: There are a few ways you can do that in SPSS; the quickest and dirtiest way if you have the Custom Tables utility is to create a nested table with each of those three variables (drag Age, Gender, and Income out into the table, and arrange them however you like), and set the summary statistic to Table % (it might be called something else, I don't have it in front of me). Then you can say females from age 35-45 making $40,000 a year or more make up 6% of the surveyed population.

*Bring traffic to the website: you can tell them how frequently those target customers use different categories of online media (email, jobs, etc.) which will help them come up with an initial display advertising plan. In SPSS, you can filter the data (go to Select Cases and set an If condition matching your target customer), and then create a table with just the Frequency of Use questions. Since I'm guessing frequency is given on a range, and it will be the same for all questions, you can put the questions in the columns and have the response choices shown in the rows, for a nice readable table. The summary statistic here will be column %.

*Sell them something: You've got an awful lot of variables about what's important and not important for online shoppers and non-shoppers, and I'm not sure I understand how it's all set up. But there is probably some sort of way to come up with a score - if they have used a rating scale of some sort, then you can use the Mean as a summary statistic, or summarize by grouping response options into groups (like 0-3, 4-6, 7-10) and reporting on some group, like the percentage that rated it 7-10. You can do this grouping in the Categories dialog inside Tables. When you have that score for each variable, you can put them in a ranked list and say "these are the 5 most important things", and "these are the 5 least important things". 



The key always goes back to starting with a good objective. Analyzing the data without an end in mind is a fruitless exercise.
A: This data set seems a tipical marketing data set. Moreover the question is very vague and obviously, the possible answers are many. I suppose you wish to investigate what are the key factors that influence variables 8 and 9 (intensity of internet purchases). I would do some descriptives about the demographic variables, then I would perform a factor analysis on items  9 to 45 and I would use extracted factors to predict 8 and 9. 
