One-Way Analysis    Targeting    Two-Way Analysis    Regression

Almost all program planning involves a stage of identifying who the "worse-off" people are, and knowing where they are located.
Ranking populations by levels of food insecurity, mortality, access to services, or environmental factors, is a device used to determine
the "worst-off" population. Mapping often helps to display such results. Ranking involves straightforward one-way analysis, showing the
outcome indicator by one classifying variable at a time, for example: prevalence of eating < 3 meals per day by district.

Breaking down the national situation analysis by region, district, etc., gives exactly the sort of information required, with the proviso that
some further selection of outcome indicators may be needed. At an area level you can’t easily use 10 or 20 indicators, as is typically and
usefully presented at the national level.

Targeting often uses ranking analysis, where it provides a comparison between geographic areas (or other grouping) for an outcome variable
ranked by prevalence, by access to services, or some other factor of interest. This can help in program planning to assign the interventions to
the areas with the worst outcome and the least current access to assistance.

Ranking Analysis

With SPSS it is possible to sort your outcome variable, such as prevalence of eating < 3 meals per day, in descending order to see which
districts are worse off. This will give you a quick look at possible relationships and give you ideas for further analysis. There are a few
methods one can use to get this information - setting up an aggregate data set, running a simple one-way association (compare means)
routine by district and independent variable of interest, and creating case summaries.

1.  Take a look at how to create multi-indicator descriptive analysis or ranking analysis for Somalia data by using an aggregate
data set.
                        The aggregate data set is now sorted by district, highest to lowest according to food sufficiency status.
                        It should look like this:

From this ranking we can also see possible associations between food sufficiency and other food and agricultural variables (although there are
better ways to do this which we will discuss later). For example, meals per day is highest for the districts with the highest mean food
sufficiency, which is a finding that makes sense. Something that isn’t quite clear is how the district with the greatest mean total area has the
lowest mean food sufficiency. One must remember, though, that these are means for the entire district, which includes anywhere from
15 – 93 cases (n_break). Thus, a small number of land owners with large tracts of land could be driving the average up. Having knowledge
of the area though, one could make an educated guess that these landowners are pastoralists, and thus need more land for their animals to
graze. Nonetheless, further investigation should be carried out in order to determine what is responsible for this finding. 

2.One step is to investigate if foodstuff, meals, chmeal, admeal, lit etc. (dependent variables) are affected
by Somalian district (independent).                              
Using compare means is an easier method in that you do not have to create a new aggregate data set, but unfortunately it does not allow
you to rank by district. The output should look like this:

                   3.         Try using the Case Studies option with the Banngladesh data set:

                    1. Open bdeshc.sav

                    2. Under the Data menu, go to Sort Cases.

                    3. Select the variable prevalence low ac in males and females (acprvfm)

                    4. On Sort Order click Descending.

                    5. Click on OK.

                    6. Click on Analyze, Reports, then Case Summaries.

                    7. Select district, acprvfm (Prevalence low arm circumference), lit (% 7+ years school ), meas (% measles immunization),
                        and roof (% with brick or tin roof).

                    8. Click on Statistics: Cell Statistics should be empty; click Continue.

                    9. Next to "limit cases to first", type in 20, and check the box to the left.

                    10. Click on OK.

Your SPSS output should look like this:

This case summary shows which districts are worst off for prevalence of low arm circumference (both males and females), as well as their
corresponding prevalence of literacy, measles immunization, and good roofing. As can be seen, it lists the 20 worst districts; in order to list
all districts simply do not limit cases (computing step # 9).