assessing for variation - simple

[Rose Matthews]

hello, I am trying to prove that low income individuals are more likely to use a variety of cars compare to high income individuals who are morelikely to use one type of car

Would you kindly recommend how I can show this?

I can use a simple

tab type car_model

Can you recommend any other way, not sure if I'm being clear?


A simple

* Example generated by -dataex-. For more info, type help dataex
clear
input float(type car_model id)
1 1 12
1 1 13
1 2 17
2 1 16
2 2 19
2 3 20
2 4 21
end
label values type m
label def m 1 "high income", modify
label def m 2 "low income", modify
label values car_model q
label def q 1 "volvo", modify
label def q 2 "fiat", modify
label def q 3 "mercedes", modify
label def q 4 "renault", modify

[Carlo Lazzaro]

Rose:
you may want to try:

 tabulate car_model type , chi2 col

+-------------------+
| Key               |
|-------------------|
|     frequency     |
| column percentage |
+-------------------+

           |         type
 car_model | high inco  low incom |     Total
-----------+----------------------+----------
     volvo |         2          1 |         3 
           |     66.67      25.00 |     42.86 
-----------+----------------------+----------
      fiat |         1          1 |         2 
           |     33.33      25.00 |     28.57 
-----------+----------------------+----------
  mercedes |         0          1 |         1 
           |      0.00      25.00 |     14.29 
-----------+----------------------+----------
   renault |         0          1 |         1 
           |      0.00      25.00 |     14.29 
-----------+----------------------+----------
     Total |         3          4 |         7 
           |    100.00     100.00 |    100.00 

          Pearson chi2(3) =   2.2361   Pr = 0.525

.

or -mlogit-.

[Denise Vella]

Try Chi squared test

[Rose Matthews]

Grazzie Carlo

however, I wasn’t really looking for significance / chi squared (i thought I should have added it in my first post)

i want to say that high income indiviuals are more likely to have one good car compared to low income who are more likely to have a variety of cheap cars.

but it can’t seem to come to me what technique I should use or in that case how to set up my data ?

another way I thought of is categorising for eg 1 car vs 2 or more cars. However, this would mean I know that the upper threshold of variety which in this case I’m saying its 2 or more

i’m sure it’s simple but I can’t seem to think how to set it up my date

[Carlo Lazzaro]

Rose:
see -mlogit, then.