Family Planning and Women's Lives in Rural China

International Family Planning Perspectives, Jun 2004 by Hardee, Karen, Xie, Zhenming, Gu, Baochang

Overall, 4,501 women and 1,526 men were interviewed in the survey; 32 women and 232 men participated in focus groups that ranged from five to 12 people. The focus groups were conducted with women and men of reproductive age, women and men past reproductive age, unmarried women and male and female entrepreneurs (a group of special interest in China). The fieldwork in Jiangsu and Anhui was carried out in 1996; because of funding issues, fieldwork in Yunnan took place in 1998. Given that the family planning policies and the social and economic situation in the three study sites remained the same during this time, the difference in the timing of the fieldwork should not affect the results.

CPIRC and the provincial and local family planning commissions chose two counties for the study from each of the three study provinces. Within each county, 10 townships were randomly selected for the survey and seven townships were randomly selected for the focus group discussions. CPIRC randomly selected one administration village (which includes several natural villages) from each township, with the requirement that the village be at least three kilometers away from a town. From each village, 120 households were randomly selected from lists of couples of reproductive age maintained by local family planning workers.

During the field survey, CPIRC interviewers were permitted to visit only the households on the list. If no one was home when the interviewer visited, the interviewer was instructed to go to the next house on the list. When 75 evermarried women and 25 ever-married men from the list had been interviewed, the survey in the village was considered finished. If a sufficient number of interviews could not be conducted in a village, the interviewer was instructed to go to a nearby village to complete the interviews. Although response rales were not calculated, the people approached in the eligible households were very likely to agree to participate; thus, overall response rates were very high.

In one county in Jiangsu, interviewers surveyed some men and women at work instead of in their homes, because most couples worked in factories during the day. More respondents in that county than in others had to be substituted; thus, the sample from that county became a convenience sample. The sample is slightly older in that county than in others because many of the younger women were working in the township. Still, most of the women interviewed in the county had one child-the same number women not interviewed likely had.10 Thus, this difference in sampling had little effect on the analysis.

For the focus group discussions, local leaders helped select participants after discussions with the research team on the desired characteristics of participants. The groups, which met in various public places, lasted 1-2 hours and were moderated by CPIRC staff. Separate sessions were held with women and men, and with different age-groups. Although it would have been ideal for the moderator to be similar to the group, particularly in terms of sex and age, this was not always possible. In addition, a local representative was present to translate, if necessary. Some groups were observed by officials from the township or province, but these observers were not known by the participants.