Table 5 Potential reasons for observed differences in the numbers of people endorsing a particular theme

The IFG participants differed between countries in terms of recruitment sources and/or sample characteristics

• Are there any systematic differences in sample characteristics between the two countries?

• Are the sampling differences a result of cultural differences in the larger population or are they due to differences in recruitment approaches?

• Do the different coding frequencies make sense based on known sample composition?

IFG moderators followed different lines of qualitative inquiry to gather information

• Were there differences in the numbers and types of probes used by moderators for the particular topic?

• Were there differences in the number and types of supplemental questions asked from other sources (e.g., session observers)?

The Coding Schedule was applied in different ways by the moderators

• Did the moderators apply different coding categories to a particular type of response? If so, what was the reasoning behind their approach to coding?

• Did existing or newly created categories overlap with other coding categories?

The observed differences might be due to cultural differences

• Did the observed frequency differences between countries result from differences in the ways respondents understood or described their condition?

• Did the ways respondents behaved or coped with their condition differ significantly?

• When discussing all the possible reasons for the observed differences, did cultural or social factors seem plausible?