responses and behaviors within case studies. For example, people may respond

differently to questions in an interview or questionnaire because they may try to give the

researcher a response he/she wants to hear. I tried to reduce artificiality by encouraging

that participants interact in a setting comfortable to the participant- non experimental

conditions. Sensitization is an uncontrollable threat. Inclusion in the study may alter

responses.

Sample Framework and Sample Selection

 The theoretical population of this study is American Indian farmers in the United

States. Farmers are individuals who actively cultivate land for food production. This

definition does not consider farm size, amount of time spent farming or value of sales. I

used this definition in order to include individuals in this study who are either cash crop

producers or home gardeners. All cases include American Indian farmers or home

gardeners who cultivated land within the study site and had the same tribal affiliation.

Many American Indian Nations/Tribes in the U.S. are from distinct traditions, languages,

and historical backgrounds. American Indian Nations are not really a single ethnic group,

but rather a collection of many ethnic groups (Champagnes, 1999). I restrict my study to

a single federally recognized tribe within the United States that has tribal members who

currently farm or garden since the phenomena under study has a strong cultural

component.

 I contacted 23 federally recognized tribes recommended by a key informant

working with Native Women in Agriculture to select a tribe within the USA. Ten tribes

of the 23 possessed cases which fit the parameters of the sample selection. Many of the

other tribes did not have many or any farmers but instead had ranchers. Two of the ten

tribes officially rejected the proposal to conduct research with tribal members. Seven