Agriculture Reference
In-Depth Information
unobserved paying attention to organic label, and as before, the researchers observe an
ordered level of paying attention, as follows:
*
ORG
_
LABEL
1
if
ORG
_
LABEL
i
i
1
*
ORG
_
LABEL
2
if
ORG
_
LABEL
i
1
i
2
*
ORG
_
LABEL
3
if
ORG
_
LABEL
(6)
i
2
i
3
*
ORG
_
LABEL
4
if
ORG
_
LABEL
i
3
i
4
*
ORG
_
LABEL
5
if
ORG
_
LABEL
j
5
i
where
I
are the unknown threshold parameters to be estimated. The first threshold
parameter is normalized to zero (
1
= 0).
To estimate the three equations [1], [3] and [5], we have assumed that the error terms (u
i,
ε
i
and
i
) may be correlated. The, instead of estimating them independently, they are
considered to be a multivariate limited-dependent-variable model, in which the three error
terms (u
i
ε
i
and
i
) follow a multivariate normal distribution with mean zero and variance
and covariance matrix
.
The limited dependent-variable (LDV) model with correlated error
terms are estimated using Hajivassiliou and McFadden's (1998) procedure implemented in
Proc QLIM in the SAS
©
9.1 statistical software package.
4.2 Data gathering and variables definitions
Data were collected from a survey conducted in Italy during November-December 2008. The
capital Rome was chosen for two reasons. The city can be considered a representative
sample of Italy since its economic indicators, such as average expenditure on food products
(498 €), average household income per year (34,000 €) and most demographic characteristics
are very close to the economic indicators for Italy (ISTAT, 2008). Sample size was set at 380.
As the population can be considered infinite, this sample size results in a sampling error of
±5%, assuming a confidence level of 95.5% (k=2) and p=0.5. A stratified random sample of
consumers was made on the basis of town district and age. A number of representative
grocery stores and supermarkets were selected in each town district, and food shoppers
were randomly selected outside these food outlets. Target respondents were the primary
food buyers in the household, and interviews were carried out face to face. Interviewers
approached the randomly selected individuals asking them one screening question, whether
they were the main household food shopper. The questionnaire was designed to analyze
consumers' organic food knowledge, attitudes and purchase behavior.
The questionnaire format was validated using a pilot survey. The questionnaire also
contained questions on socio-demographic characteristics (i.e. sex, family size and
composition, age, education level, income) and lifestyles. Summary statistics for the
characteristics of the full sample are presented in Table 2. The majority of participants are
female (58%) and, on average, participants are living in households of 3 members. In
addition, the average age is about 45 years. Moreover, around 15 % belong to high income
groups and about 48% of the subjects have a university degree.
The first question was related to their knowledge of organic food products. The level of self-
reported knowledge was measured asking respondent their self-reported level of
knowledge from 1 to 3, where 3 indicate the highest level of knowledge (KNOW). Objective
Search WWH ::
Custom Search