Population Estimation assignment

Assignment 1: Population Estimation Exercise

Objectives

Large scale population displacements require responders to rapidly assess an often chaotic situation and plan for immediate provision of basic needs such as shelter, food, water and sanitation.  In order for your organization to adequately plan humanitarian assistance, you have been asked to provide the following information about the displaced population:

• total number of households
• total population size
• average household size
• estimate the number and percent of vulnerable households, including:
  1. Child headed households
  2. Female headed households

This exercise is based on an in-class simulation using a population of poker chips.  Poker chips represent households (or in the case of suited poker chips, household characteristics) and your task is to estimate population size based on a sample of poker chips.

Methods

To estimate population size, it has been recommended that you count households in a sample of locations and then extrapolate these findings to the entire population. Consider the following guidelines:

• 30 clusters of different household sizes were selected using systematic random sampling method considering that the distribution of the population throughout the camp is relatively homogeneous.
• The wall panels in Feinstone Hall (see sample planning sheet) can serve as the basis for a sampling grid. Using this approach, a 12 x 18 grid of squares (3.3 ft x 3.3 ft) each can be superimposed on the room.   For ease of extrapolation, each area sampled (circle or square) covers the same area as a square in the grid and is equal to 10.9 sq ft or 1/216 of the total area of Feinstone Hall.

Steps to Completing the Assignment

The Population to be Counted

We recommend entering chip counts for each location into an Excel spreadsheet (where each location is a row and chip colors/types are column headers). Further analysis can be done using excel, stata or other statistical packages.  We recommend the following steps for determining the requested information.

1. Tabulation of findings:
   1. Determine the number of households, population size, and average household size in each cluster.
   2. Determine the proportion of female and child headed households in each cluster
2. Sample statistics: determine the median and the mean and 95% CI for each of the following:
   1. Number of households per cluster
   2. Total population per cluster
   3. Household size
   4. Number and proportion of female and child headed households per cluster
3. Using the information from step two, extrapolate your findings to determine estimates for the entire population.

Assignment 1: Population Estimation Exercise

Present the final results of the population estimation exercise as if you were writing a journal article.  Include the sections and information outlined below.  Your submission should include no more than one table of results and one page of text (font 11 or larger, single spaced, margins one inch or greater).

1. Methods: describe the sampling method used, any decision rules that were developed and any differences between the planned and actual sample. Include a Feinstone Hall diagram that illustrates your sample with the assignment. (4 pts)

1. Results: present the data collected in the exercise as you would if you were preparing a journal article.  Consider what information should be presented in table format versus text.  Include sample descriptive statistics for household size, the total number of households per sampling unit, the number of female and child headed households per sampling unit and the total population per sampling unit.  Also provide point estimates and confidence intervals for the total number of households, number of female and child headed households and the total population size of the Feinstone Hall poker chip population. (12 pts)

1. Discussion: describe any limitations to the methods used to derive population estimates and (dis)advantages to the sampling method used.  If applicable, discuss any concerns you have about the population estimates and reasons for why they may over or under estimate total population size. (4 pts)