Comparison of two question sequences for assessing pregnancy intentions

Area of Science:

• Reproductive Health
• Survey Methodology
• Sociology

Background:

• Unintended pregnancies carry significant health, social, and economic consequences.
• Pregnancy intendedness is categorized as intended, mistimed, or unwanted.
• Survey question wording and order can influence respondent understanding and data validity.

Purpose of the Study:

• To assess the impact of question order on reporting of pregnancy intendedness.
• To compare the National Survey of Family Growth (NSFG) and Demographic and Health Survey (DHS) intendedness questions.
• To identify factors associated with discordant responses in pregnancy intendedness surveys.

Main Methods:

• A randomized crossover design was employed in a 1993 survey of Arizona women aged 18-44.
• Participants responded to both NSFG and DHS intendedness questions.
• Demographic and pregnancy-related factors were analyzed for their association with discordant responses.

Main Results:

• 25% of 2,352 ever-pregnant respondents provided discordant answers between DHS and NSFG questions.
• Age, marital status, income, education, parity, time since pregnancy, and pregnancy outcome predicted discordant responses.
• Young, unmarried women were more likely to classify pregnancies as 'mistimed' when the question was asked later.

Conclusions:

• The classification of pregnancies as intended, mistimed, or unwanted can be challenging for women uncertain about future reproductive preferences.
• Survey validity can be improved by addressing respondent ambivalence and clarifying definitions.
• For young, unmarried women, avoiding classification of unintended pregnancies as mistimed or unwanted may enhance data accuracy.