Findings, implications and lessons learned

  • Although stratification was used in this survey, we have deliberately ignored it so as not to overcomplicate the calculations.

  • It was observed that the distribution of cattle per homestead was skewed. It is possible that a logarithmic transformation might be applied to the raw data in order to improve the precision of the population estimate. One of the study questions addresses this.

  • Two different formulae have been used for calculating standard errors of population totals:
      ST = √{[(T1 - T)2 + ... + (Tn - T)2] / (n-1)}
      ST = √{[(T1 -M1)2 + ...+ (Tn -Mn)2] / (n-1)}.

    We have applied the first formula to Scenario 1 (equal probability) and the second to Scenario 2 (probability proportional to dip-tank area size).