impartial. As outsiders they may make critical judgements that insiders might avoid. All organisations should undergo such a process from time to time. Indicators developed by an institution may also assist its partners or external evaluators to understand the issues that the NHRI considers most important. A few examples will illustrate how data can be (mis)interpreted. National institutions that responded to our questionnaire indicated that the ­ indicator most frequently used was the number of complaints received, ­supplemented by information on types of complaints and rates of ­ resolution. It was also clear that most NHRIs actually measure outputs rather than ­performance or impact. Reports might list the number of press briefings issued, but not the impact that these outputs had on media coverage or whether media coverage secured the NHRI’s objectives. With respect to statistics on complaints, national institutions generally make the following assumptions: ▪ The higher the proportion of resolved cases to the total number of cases received, the better their performance. ▪ An increasing number of new complaints is evidence of greater human rights awareness, and thus of public awareness campaigns’ success. These assumptions may of course be correct in any given instance. As we have seen, however, such statistics may be evidence of quite different trends. If ­complaint rates increase, for example, the common-sense interpretation is that human rights violations are increasing. This is not inconsistent of course with the interpretation that there is a greater popular awareness of human rights. ­Equally, if complaints decline, does it imply that the NHRI’s public awareness work is failing? Or that human rights violations are being combated ­successfully? It is impossible to know the true answer without supplementary information. The point is that these interpretations can only be validated using other ­information, such as an assessment of the overall human rights situation, interviews with complainants to discover how they learnt about the NHRI complaints mechanism (based on qualitative participatory indicators), and so on. The conclusion that a high proportion of resolved cases indicates success seems at first sight to be less controversial. Yet it relies on the further ­assumption that resolving complaints is the best way to deal with human rights problems. This too can be challenged. First, resolving complaints is essentially a method that NHRIs borrowed from Ombudsman offices, whose function is to deal with maladministration – ­delays or failure to deliver information or a service. Resolution of such cases is ­neither ­ adversarial nor punitive; almost invariably, it consists of providing the ­complainant with the service to which the latter was originally entitled, often accompanied by an apology. By contrast, human rights violations are often crimes, for which those responsible should be held accountable. They may also be wrongs in civil law. Either way, the victim, and society at large, are entitled to redress. Blame-free resolution is not sufficient. Assessing the Effectiveness of National Human Rights Institutions 29