Page 237 - Contributed Paper Session (CPS) - Volume 7

CPS2073 Alban Çela et al.

Calibration for the Census of Agriculture in
Albania
2
Alban Çela , Klajd Shuka
1
1 Agriculture and Environment Statistics Director at Institute of Statistics Albania
2 Finance and Supporting Services Director at Institute of Statistics Albania

Abstract
Agriculture is one of the main economic activities, impacting gross domestic
product of the country, contributing with around 20 percent. Accurate
agriculture indicators are crucial for the European integration of Albania. The
Institute of Statistics in Albania is conducting every 10 years the Census of
Agriculture. This census is one large scale statistical activity for the collection,
processing and publication of data on the structure of agricultural holdings.
The 2012 census of agriculture data comparative analysis showed deviations
from the historic administrative data. This paper presents calibration as an
alternative method to estimate agriculture figures, based on census of
agriculture data and administrative data. Fisher Ideal Index is used for
calculation of trend components. This method adjusted the figures for
different type of bias such as under-coverage and non-response. As a result
of this analysis calibration process to improve the quality and accuracy of
agriculture data that will be used for the creation of the statistical Farm
Register.

Keywords
Estimation; Calibration; Bias; Administrative data

1. Introduction
INSTAT is the leader of the National Statistical System and main producer
of official statistics in Albania. During the implementation of official statistics
programmes many surveys and censuses have been conducted. Usually
censuses in Albania are conducted every ten years. For censuses usually the
indicators are produced based on the collected data taking into consideration
the effects of under-coverage. On the other hand, for surveys estimations are
made taking into consideration under-coverage and different types of non-
response. To reduce the negative impact of the previously mentioned
problems the estimates are adjusted during weights calculation and
calibration phase. For the Census of Agriculture (CA) of 2012 for the first time
INSTAT has been calibrating census data, due to differences with other
administrative historical data. High quality and accurate data from CA are the
basis for the creation of the statistical Farm Register and have a big impact on
the quality of indicators produced using future surveys on agriculture.
224 | I S I W S C 2 0 1 9