|
Sampling Plan and Methodology --------------------------------
|
|
|
| Establishment Survey |
| |
| The
Establishment Survey is conducted to understand the demographic
profile of the market. Hence aMap ensures that
it is large enough to capture the smallest nuances of
the market. aMap also ensures that at least 50%
of the total fieldwork done is checked for proper recording
of answers & the authenticity of interviews taken.
|
| |
| In a city like Mumbai, the
total sample size of the Establishment Survey was 10,000
interviews spread across 1,000 clusters. The selection
of cluster head addresses is done using the Electoral
Rolls as the sampling frame. |
| |
| aMap uses 2001 census population
figures to arrive at the current estimated population
and conducts its own Establishment Survey, on the basis
of which the Panel composition is derived. |
| |
|
| aMap also uses this survey
as the sampling frame for selection of panel homes. |
| |
| Panel Controls |
| |
| Panel Controls are
the parameters on the basis of which the panel is constructed.
These are broadly grouped into two areas: |
| |
| Primary Control - These
parameters are used to weigh the data. |
| |
Secondary Control -
These parameters are important during panel construction. Representation
of these ensures that the data is not skewed e.g. The tendency
of a home surfing across channels is higher in homes, which
have a TV remote.
It is therefore important that the representation of these homes
is proportionate to the universe.
|
| The control parameters
are: |
| |
| Primary
Controls |
Secondary Controls |
| |
- SEC
- Terrestrial/C&S
- Age
- Gender
|
- Family Size
- Language Spoken at Home
- Geographical Dispersion
- TV type (B&W / Color)
- Remote / Non-Remote
|
|
Additionally, special efforts are undertaken to
ensure SEC A+ presence in the panel. |
|
|
| |
|
|
| |
|
Each
meter is connected to a GSM modem. There are three time windows
during which central server dials into each meter to collect
data. In case the data collection from a meter fails in the
first time window, two more attempts are made in the next
two windows.
|
| |
|
|
| |
|
Twenty
percent of panel homes are replaced with new homes every year.
Sample rotation activity is distributed across the year.
|
| |
