Selecting the right dimensions for flawless tables


We are proud to launch the first blog of our new team of bloggers today. The first blog being from Danielle Moore, M&E manager of AIP Foundation (Asia Injury Prevention Foundation). Danielle is currently based in Vietnam, where she is leading the roll out of ActivityInfo in various countries for the Asia Injury Prevention Foundation.

Selecting the right dimensions for flawless tables

The reporting function in ActivityInfo brings the data management process full circle. You have designed your database, you have trained your team to enter data, and now you are ready to report to your donor. While charts and maps may be effective visual representations of your data, tables can be the most flexible option for extracting the results from your project, especially across multiple sites and activities.

Choosing the right dimensions can ensure that your data is displayed accurately. If you are new to using ActivityInfo, you may find the wide selection of dimensions a bit overwhelming. To help you and your team create flawless tables, here are the some of the most useful dimensions and when to choose each.

1. Indicator

If your table includes multiple indicators in units that cannot be aggregated or that you wish to examine in further detail, you should choose indicator as a dimension. Indicators can currently not be accurately aggregated when some are numbers and others percentages and when they measure different units.

For example, we have a table that reports the number of people surveyed (50), the number reporting awareness of a new law (22), and the percentage reporting awareness of a new law (44%). If indicator is not included as a dimension, the table would calculate the sum (116) of both the numbers and the percentage, which has no meaning.

In a difference case, say you are reporting on volunteer involvement in your event. You choose the indicators, number of volunteers (5) and number of volunteer hours (20). It may seem obvious, but in all the choices we make to set up our table, you could easily forget to choose indicator as a dimension and end up with the total of the people and the hours (25), which does not accurately describe the volunteers’ engagement.

2. Site

The site dimension is necessary when you are reporting data for multiple sites in percentage form. If the sites’ populations vary, as they do in most cases, you should weigh the average of your percentages according to the number of people.

For example, if you are reporting the average score (%) of students on a knowledge assessment at three schools, you should select the number of students at each school and the average score at each school as your indicators, then you can calculate the weighted average according the the size of each student body.

3. Form (Previously Activity)

If you have the same indicator(s) from multiple forms in your table, consider making form a dimension instead of time. This could help you identify what you did to generate those outputs or outcomes.

My team uses form as a dimension when we are reporting media coverage of our project. Each of our forms, such as public events and schools activities, includes an indicator for number of press received, so we would choose each of those indicators for our table. Then by selecting form as a dimension, we can see which activities generated the most media. Especially for projects aiming to influence public behaviour and for donors hoping to improve their media presence, this information could help influence future project design.

4. Attributes

Remember that there are attribute dimensions at the base of the list.  These can provide more context to your data then the generic dimensions.

For example, we have a table reporting the results of monitoring, such as behaviour observations, that takes place at several intervals throughout the project. An attribute dimension, such as stage, that indicates whether the observation took place pre- or post-intervention positions the change in behaviour in relation to an event, helping to establish a causal relationship between your project and your outcomes.

This is not an exhaustive list. Maybe you want to see how attendance at your clinic fluctuates throughout the year, then select the month dimension under the time drop-down. If you entered targets for treatments per month, choose the Realized / Targeted dimension to see how you have progressed toward your goal.

To conclude

So ActivityInfo pros, what have I missed? What dimensions are most useful for your tables? How do they help you adapt your project as you go and evaluate your effectiveness?