What to Collect
Three Types Of Data
- Cross-sectional. These are data collected at a single point in time from a cross section of a population, giving an impression of the status of the studied groups based on a sample.
- Time series. The same variables examined in a cross-sectional study may be collected over a period of time with repeated samplings, usually at regular intervals. Studying time series can be very informative about change processes, but typically a long series of observations is needed (upwards of 20 time points and preferably more than 50) for formal statistical time series analysis. Shorter time series can still be very informative but be aware of the dangers of over analyzing short series.
- Pooled (also called Panel). Pooled (or panel) data are cross-sectional data collected over time, e.g. production and input cost data for a sample of cotton growers in Arizona (cross-section) collected annually for 5 years (series).
The First Step To Economic Analysis
Choose the level of analysis you want and also be sensible about the quantity and quality of data you can expect from your participants.
- The level of analysis and quantity of data you can expect will dictate the answer to questions about what to collect. For example your preference might be to calculate partial budget estimates of net gains for groups who have and have not adopted an IPM practice. However if you know that many of the participants will be unlikely to have the information needed for a partial budget analysis, you might still be able to calculate cost-effectiveness for the IPM practice in comparison with the "standard" practice.
- Remember that the best quality analysis will result from a study in which time and effort have been invested in thinking about the data collection well in advance. Working within the framework of a Logic Model will help you focus on what you want to know and which data you need. Generally, whatever the context, questions should be as specific as possible. The information you attempt to collect should be something which is measurable (a pilot study or focus group with participants in advance of the main project can help to verify this). Finally, to make sure that you complete the impact assessment on time, you should establish a timeline for completion at the outset.
- Keep in mind the extent to which what is feasible may be limited by available resources and don't over promise. Also don't forget that in some circumstances, certain pre-requisites may be prescribed by the terms of the funding for the work (e.g., you may be obliged to carry out a cost-benefit analysis whether or not you think this is the best option)
Examples Of The Types Of Data To Collect
- Pesticide costs
- Scouting / monitoring costs
- Prices and yields (agriculture)
- Overall pest control costs (such as costs of traps, beneficial insectaries, biocontrol, resistant hosts, avoidance strategies)
- Health-related costs (such as absenteeism, medical costs, mental health)
- Environment-related costs (such as decontamination, clean up, endangered species, loss of species diversity)
- Levels or densities of pests, diseases or weeds
- Amounts of crop damage, yield loss, or livestock health costs
- Health costs to consumers, local community or laborers
- Recreational or aesthetic costs (quality of life, mental health, tourism)
Non-agricultural/Structural Settings (e.g. schools, housing, or community IPM, Housing IPM):
- Costs for pesticides and pesticide devices used
- Pest Management professional contracts
- Costs for employees that work for institution (such as in-house pest control, could also include custodial, food service, maintenance)
- Overall pest control costs (such as pest proofing materials, sanitation, clutter removal)
- Pest infestation levels
- Health costs (such as absenteeism, medical, allergen, mental health)
- Environmental costs (such as decontamination, clean up, water quality)
- Pesticide exposure indices for employees, school attendees, householders
- Environmental pesticide exposure indices