prev next front |1 |2 |3 |4 |5 |6 |7 |8 |9 |10 |11 |12 |13 |14 |15 |16 |17 |18 |review
So how do you go about selecting your study type?

It obviously depends on what question you are trying to answer, whether you are lucky enough to find a source that covers your target population, and whether you are allowed access to the data. Alternatively, you may find that the group you are after is homeless people and they are not listed anywhere and are not contactable on the phone or may use different names on various registers.

For some topics (e.g. attempted suicide, or sexual behaviours) you may be restricted in the type of population you can approach, even if the event happens across all genders, socio-economic class and age-groups. For example, older people would be less likely to participate in a survey addressing those sensitive topics. Likewise, young people may not be interested in responding to questions about health beliefs and diabetes, which are not exciting and require some knowledge.

Finally (and sometimes this is the determining factor) you may have designed a grand survey and have access to the sampling frame but your resources or your deadline for delivering the report do not allow you to use good techniques or good data sources. In which case I would recommend you undertake a pilot study or do not conduct the study at all.