When working on a subsequent wave of a study we need to know if data position matching is required between the current and the previous wave. Since a wave is a survey that was already programmed and is being fielded again after a period of time, either without changes or with changes, knowing if matching the data fields is needed or not is required up front. If data matching is not required the team doesn’t need to worry about matching the previous data positions.
Many of our clients use the same analysis setup from wave to wave. Because of this, if the positions change this can throw off existing calculations, formulas, etc. To help with this, we can keep deleted positions in the data file and add any new positions to the end. This applies to ascii and csv and is usually not needed for SPSS since we will keep the existing data labels the same as long as it is feasible. If a question changes (i.e from a single select to a multi-select) than we will need to add a new question to account for this so the labeling will change.
It typically does take longer to match positions so it can incur additional costs, however we will always do our best to match the labels so this is included standard.
Matching positions would be quoted but matching labels is standard.
There are three different types of data matching typically requested by our clients. Here are the definitions for each type:
- Variable Name Matching – Maintaining variable names of pre-existing questions is our standard if we are programming a study as a copy of a previous version we programmed. If questions are removed those data fields will be removed unless otherwise requested. If we are programming the survey as new, we encourage our clients to update the document to reflect the labels they want in the survey. If the survey is marked with the desired labels, we can make the updates during the programming phase. If not, than we would encourage use of our Layout Mapping Template.
- Description Matching – The description is the “Label” in a standard SPSS data file. This uses the text programmed into the survey. If this needs to be matched to a separate layout on a new study, we would recommend using the Layout Mapping Template. If this is a copy of a study programmed by Jibunu it will only change if the text within that question changes.
- Position Matching –The position is the order the variables appear in the data file. If we are matching positions, items from previous waves are not removed from the layout (they will be hidden in programming instead of deleted) and new items are moved to the end of the file. We will need to know which layouts need to be matched (CSV or ASCII). Generally, this is not needed for SPSS, but please check with your data processor if you are not sure what they will be using.
Layout Mapping Template
If you previously used a different vendor on a tracker or wave study, but are concerned about maintaining your data file, this product will allow you to simplify the data conversion process and reduce the costs that are usually associated with this. Once the survey programming is final and approved, we will send you a pre-populated layout mapping template for the survey with detailed instructions on how to fill in the preferred text. This template can be used to change all or one of the following: Names, Labels, Value Labels and Positions.
