Reduction in the number of records

[sergiofurtado]

Hello everyone,

Considering a situation where we have multiple records (eg 90 types of foods) and each of these records has 15 items (food characteristics) and these records are directed to some people in a household (the application would have residence data and of your people).

The simplest strategy would be to create one record for each type of food, ie create 90 records.

Is there any other strategy to reduce this number of records while still collecting all the food data?

Thank you very much in advance

[Gregory Martin]

Another option is to only collect records for the foods that people consumed (or purchased, or whatever you're asking). In that way, on every new occurrence, you would display a dynamic value set showing all of the foods that haven't been selected up to that point, and then that occurrence of data collects data for that food.

However, is there a reason that you are trying to reduce the number of records? If you're concerned about file size, the space taken by a mostly blank record is pretty minimal.

[sergiofurtado]

Hi Gregory,

My concern is if an application with a very high number of records has any implications during data collection or after data export. These records are required even if the randomly selected resident has not consumed a particular food, for example recording a frequency of zero.

On the other hand, it will take a long time to structure the questionnaire and this strategy will also involve the collection time.

You may be worrying me too much about this, so decide to consult them on another approach that could improve the structure of the application.