4.3 Limitations
The research activities were facilitated in a number of ways:
first, the dry season and absence of significant hurricanes allowed easy access
to all areas. The security situa- tion in Cap-Haitien was good enough to allow
the researcher to go in all areas without restrictions, except at night,
contrary to other cities like Port-au-Prince in which some areas are forbidden
to foreigners and NGO staff due to high levels of violence. Oxfam had developed
good relationships with the local population, most notably by setting up about
20 KLPS (local committees for civil protection). The language barrier was not
high as Haitian Creole is derived from French, and the researcher learnt enough
Creole
to be able to ask most questions and understand their answers
by himself. Yet, what most facilitated the research was the presence of an
assistant, Guettie Noël, who acted both as a translator and facilitator
during fieldwork, who had experience in research, who worked for the Ministry
of Public Health and had good knowledge on sanitation. She also attenuated the
possible gender bias as the researcher is male; finally, she had a
«pro-poor» attitude which helped to reach the poorest during visits
and interviews, and hear what they had to say.
On the other hand, it is possible that the translations have
hidden some aspects to
the researcher, especially during the Focus Group Discussion as
discussions went much faster. The fact that the assistant was of Haitian
culture may have introduced a bias
as well, for instance on an occasion during the survey, when
she refused to enter a house because she had seen clear signs that it was the
house of a voodoo sorcerer. This occurred only once, however.
Other aspects constrained the research. Most notably, as the
researcher was working
for an NGO, its objectives had to fit in the project and at
some point, local Oxfam staff had requests which would not fit in the research.
Oxfam staff was interested in getting results which would be useful for the
water and sanitation project as a whole, whereas
the researcher was concentrating on excreta management. Local
staff was also keen to have the researcher work in certain areas where Oxfam
has already done other projects:
the local committees were asking for help with their water and
sanitation situation, and
the researcher was directed to areas which were not always the
most appropriate or urgent. As a result, the areas considered in the survey are
a mix of Oxfam target areas (Shada, Mansui) and non-target areas (Bas-Ravine
and Petite-Anse).
The researcher, as an outsider, was also subject to a certain
extent to the biases pro- posed by Chambers (1983): the «spatial
bias» (staying on the main road) was avoided
by going as far as possible in each zone and including remote
areas in the survey. Dur-
ing early field visits, however, limited time did not allow the
researcher to go far from
the main roads. The «project bias» (going in areas
where there have been projects) was clearly present as Oxfam directed the
objectives of the research. During fieldwork, the researcher was less submitted
to pressure. The «person bias» (meeting elite and active males) was
in part avoided with the help of the assistant; but it was striking that al-
most all representants from organisations were male, as well as most
participants in the workshop. The «professional bias» was perhaps the
most marked, as the researcher was concentrating on his topic; it was in part
compensated by being attentive to other com- ments, unrelated to sanitation,
which could also indicate that other issues could have a higher priority than
sanitation; see also the Discussion part on page 56.
Finally, the limited time added a constraint: more time would
have allowed to perform some follow-up on the ecological latrine experiment in
Mansui, to do a few more FGDs, and possibly to investigate more the
possibilities of the product-service package (see on page 51) by involving
local masons and bayakous in the process.
Rémi Kaupp
| 
