|
Probit regression: Potatoes
|
| bet | 7/25 | | Sana | 10.04.2017 | | Hajmi | 235,5 Kb. | | #4028 |
Table 5: Range of estimates from Probit model on probability a consumer is pushed out of the market for each of the GM-labeled foods. (TRY PROBIT WITHOUT POSITIVE INFORMATION)
Food product
|
Weak test – negative information
|
Strong test– negative information
| Weak test – verifiable information |
Strong test – verifiable information
|
Potatoes
|
0.73 to 1.20 ***
|
0.49 to 0.99 *
|
-0.49 to -0.71 **
|
-0.11 to -0.23
|
Tortilla Chips
|
0.36 to 0.54 *
|
0.67 to 0.98 **
|
-0.06 to -0.19
|
-0.22 to -0.48
|
Vegetable Oil
|
0.77 to 0.96 ***
|
0.70 to 0.93 *
|
-0.58 to -0.69 ***
|
-0.01 to -0.04
|
All Three Foods
|
0.71 to 1.20 **
|
0.40 to 0.76 *
|
-0.67 to -0.68 ***
|
-0.10 to -0.31
| *** The coefficients for all alternative probit model specifications are statistically significant at the 5% level (or higher) ** The coefficients for all alternative probit model specifications are statistically significant at the 10% level (or higher)
* The coefficients are statistically significant (at the 10% level or higher) for some of the probit model specifications, but not all specifications.
Appendix Table 1. 2000 Census of Population Demographic Characteristics of Polk County, IA (including Des Moines area) and Ramsey County, MN (including St. Paul area)
Variable
|
Definition
|
Polk
|
Ramsey
|
Average
|
Gender
|
1 if female
|
0.52
|
0.52
|
0.52
|
Age
|
Median age
|
45.7
|
45.7
|
45.7
|
Married
|
1 if the individual is married *
|
59.5
|
51.4
|
55.5
|
Education
|
Years of schooling **
|
13.52
|
13.76
|
13.64
|
Income
|
The median households income level (in thousands)
|
46.1
|
45.7
|
45.9
|
White
|
1 if participant is white
|
0.9
|
0.8
|
0.85
|
All variables are for individuals of all ages, except for Married, which is for individuals 18 or older, Education, which is for individuals 25 or older, and age, which is for individuals 20 or older.
* The estimate of the number of married people who are 18 or older was obtained by taking the number of people married over 15 and assuming that the number of people were married at ages 15, 16, and 17 were zero – this gives the percentage of people who are married who are 18 or older.
** The years of schooling was estimated by placing a value of 8 for those who have not completed 9th grade, 10.5 for those who have not completed high school, 12 for those who have completed high school but have had no college, 13.5 for those with some college but no degree, 14 for those with an associate’s degree, 16 for those with a bachelor’s degree, and 18 for those with a graduate or professional degree.
Appendix Table 2. Probit model: Tortilla Chips – Dependent variable = 1 if a consumer is out of the market for tortilla chips (i.e. bid = 0)
(Standard errors in parentheses, N=172)
Regressors
|
(1)
|
(2)
|
(3)
|
(4)
| (5) |
(6)
|
(7)
|
(8)
|
Intercept
|
-2.045 **
|
-1.934 **
|
-1.435 **
|
-1.931 **
|
-1.876 **
|
-1.542 **
|
-2.295 **
|
-2.196 **
|
|
(0.410)
|
(0.432)
|
(0.437)
|
(0.557)
|
(0.513)
|
(0.460)
|
(0.613)
|
(0.615)
|
Anti_info
|
0.985 **
|
0.963 **
|
0.670 *
|
0.791 *
|
0.748 *
|
0.692 *
|
0.857 **
|
0.874 **
|
|
(0.441)
|
(0.438)
|
(0.394)
|
(0.410)
|
(0.409)
|
(0.401)
|
(0.044)
|
(0.429)
| | |
|
| |
