H. Data Collection, Storage and Management

Planning for data collection includes making important decisions about the people involved and the data itself. Consider the following:

1. From Whom Will Information Be Collected?

If information from parents is desired, will data from all parents with children at a particular school be collected? If information from students is needed, will all students be surveyed or interviewed or observed? Selecting a smaller group of people to represent the entire group of interest (like parents or children) can make data collection much easier. However, it can be tricky because the appropriate size (sometimes called the sample size) may vary depending on the question. For example, if the number of children walking or bicycling to school is desired, the appropriate group from which to collect data may be the subset of students who live within two miles of school. If awareness of safety issues is to be measured, the entire student body (and their parents) may be the appropriate sample. If gathering information from the entire student body is not possible, a representative sample is needed, which may require stratified random sampling. While this term may sound complex, it only means separating students by grade, sex, race/ethnicity or another variable and randomly selecting some from each sub-group (for example, ten boys and ten girls from each grade, randomly selected as every third boy or girl who enters the classroom on data collection day).

2. Data Collectors

Consideration must be given to who will collect the data and how they will they be trained. Good training helps ensure consistent data collection from different data collectors. Without this, it is difficult or impossible to ensure that results are accurate. Common errors, such as entering the wrong information, may arise out of boredom or fatigue. Ideally, having a second person enter all the data or a fraction of it again helps to avoid this error by testing how frequently errors are occurring and working to fix the problem. If an evaluation specialist has been hired, he or she will be responsible for training data collectors and making a plan to prevent errors.

3. Protecting Confidentiality

Confidentiality is often of concern to schools and many districts have their own policies and procedures that must be followed. For example, if any data will be collected from students, informed consent from the parents is often required.

4. Data Storage

In most cases, data will be stored electronically in a spreadsheet or database. This is helpful because once raw data from paper forms (tally sheets, surveys, etc.) has been converted to a reliable electronic format it can be transferred to other software programs for analyses and/or graphic design.