|
Each country’s sanction is a unique event. In many cases, sanctions are national in scope with the result that control groups (people within the country not affected by sanctions) with which to make comparisons may be lacking.24 Control groups are the main way that difference outcomes can be attributed to a particular cause. For example, to test a new medicine, one group takes the pill while a comparable group does not (. . . or takes a placebo). Yet with sanctions, it will generally be difficult to identify an external group available for such simple, straightforward comparisons.
Comparison groups may be military versus civilians within a country, women versus men, the elderly versus adults, those receiving rationed food versus those who do not, or those employed in the public sector versus those in the private sector. The unique opportunities for generating subgroup analysis in each situation can only be determined locally. With creativity, comparison groups of some kind can be identified for many cases of sanctions. In Cuba, for example, the subgroup of the population with relatives who sent remittances from other countries was found to have better nutrition; their dollar incomes partially protected them in the 1990s from the accumulating stress of sanctions. Identification of vulnerable groups and the mechanisms by which they become vulnerable is key in determining what group to select for comparison. This process is explained in more detail in section 5.3.1.
When it is difficult to identify a control group within the country, a neighbouring country or group of countries can be compared to the effected country as a control. Of course, the validity of the comparison depends on how closely baseline conditions in neighbouring countries match those in the sanctioned State or region. For example, trends in infant mortality in Serbia were compared with similar data in neighbouring countries. Despite claims by Serbian authorities of sanctions’ harm on infants, it was found that not only did the rate of mortality decline in Serbia under sanctions, but it declined more than in any neighbouring country. There is a large body of literature on these kinds of studies, called quasiexperimental studies.25
In using neighbouring States to identify comparison groups, investigators need to be aware that “thirdparty” States in proximity to the target State (either geographically or linked economically) may also experience humanitarian impacts of sanctions on the target State.
The longitudinal studies described in section 4.3.2 above sometimes create the possibility of comparing the group that is affected by sanctions with the nonaffected (control) group.
In addition to comparisons between different population groups, changes in conditions experienced by the same population group over time can be used to assess the impact of sanctions. An example of the use of crosstime comparison for assessing the humanitarian implications of timber sanctions on Liberia is provided in box 3.
The key principle in making comparisons across time is to ensure that the analysis takes into account factors other than the variable of interest (i.e., sanctions) that may have come into play, or changed in their intensity, over the same time period.27
| 
