I know, I know.  Data analysis is not everyone’s favorite topic, but it’s a topic you can’t ignore if you want to be successful with grant writing.  Not only do you need to be able to analyze your data appropriately to accurately and effectively describe your need for the project in the needs section, but you also need to describe how you will analyze data as part of your evaluation plan.

I have read many grant evaluation plans. Most do a decent job of describing what data will be collected and how/when it will be collected.  The majority also discuss how the data will be used for program improvement purposes.  But when it comes to talking about how the data will be analyzed (one of the scoring criteria in most government grants, and many private ones, too), that’s when most grant writers fall apart.

There isn’t enough time here to discuss all of the detail you need to know regarding data analysis (hmmm….do I sense a series coming on?), but let’s start with three basic concepts in analyzing the data that you should address.

Data Collection – Like I said, most people cover this pretty well in their evaluation plans.  You need to include what data you will be collecting, how you will collect it, when you will collect it, and who will collect it.  If new instruments (surveys, etc.) are going to be developed, you’ll need to describe that process, too. Think through the whole process from developing or acquiring the instruments through getting the data into your computer for analysis.  Yes, I did say, “into your computer for analysis.”  The days of tallying surveys by hand on paper are over.  Accept it.

Descriptive Statistics – This is a fancy way of saying that you’ll use the data to describe something.  Descriptive statistics include frequency counts, percentages, means, etc. You’ll use descriptive statistics to describe the population you served.  You’ll use them to describe your basic outcome data (survey results, etc.).  Of course, whenever possible, you should disaggregate your descriptive statistics by important subgroups to make sure you painting an accurate picture. Most of the time, descriptive statistics are all you need for a basic program evaluation, but not always…..

Inferential Statistics – O.k., here’s where we separate the men from the boys….or the women from the girls…or the real evaluators from the pretenders. Inferential statistics are used to help you make judgements about the data beyond what can be said by looking at the descriptive data alone. Inferential statistics help you determine the statistical significance of the changes you see (the likelihood that the changes occurred as a result of your treatment, rather than by chance).  They help you predict things, too. If you ever studied anything beyond descriptive statistics in school, you entered the world of inferential statistics.  It’s a scary place for some, but it’s the only place to go if you really want to show causation (that your program really made a difference), and isn’t that what evaluation is all about?

If you need a refresher course on research methods, the Research Menthods Knowledge Base is a great place to start.

The GrantGoddess.com Program Evaluation Resources page has some links to interesting articles on data collection and analysis, as well as a link to two free webinars we have posted on evaluation basics.

I know, I know.  Data analysis is not everyone’s favorite topic, but it’s a topic you can’t ignore if you want to be successful with grant writing.  Not only do you need to be able to analyze your data appropriately to accurately and effectively describe your need for the project in the needs section, but you also need to describe how you will analyze data as part of your evaluation plan.

I have read many grant evaluation plans. Most do a decent job of describing what data will be collected and how/when it will be collected.  The majority also discuss how the data will be used for program improvement purposes.  But when it comes to talking about how the data will be analyzed (one of the scoring criteria in most government grants, and many private ones, too), that’s when most grant writers fall apart.

There isn’t enough time here to discuss all of the detail you need to know regarding data analysis (hmmm….do I sense a series coming on?), but let’s start with three basic concepts in analyzing the data that you should address.

Data Collection – Like I said, most people cover this pretty well in their evaluation plans.  You need to include what data you will be collecting, how you will collect it, when you will collect it, and who will collect it.  If new instruments (surveys, etc.) are going to be developed, you’ll need to describe that process, too. Think through the whole process from developing or acquiring the instruments through getting the data into your computer for analysis.  Yes, I did say, “into your computer for analysis.”  The days of tallying surveys by hand on paper are over.  Accept it.

Descriptive Statistics – This is a fancy way of saying that you’ll use the data to describe something.  Descriptive statistics include frequency counts, percentages, means, etc. You’ll use descriptive statistics to describe the population you served.  You’ll use them to describe your basic outcome data (survey results, etc.).  Of course, whenever possible, you should disaggregate your descriptive statistics by important subgroups to make sure you painting an accurate picture. Most of the time, descriptive statistics are all you need for a basic program evaluation, but not always…..

Inferential Statistics – O.k., here’s where we separate the men from the boys….or the women from the girls…or the real evaluators from the pretenders. Inferential statistics are used to help you make judgements about the data beyond what can be said by looking at the descriptive data alone. Inferential statistics help you determine the statistical significance of the changes you see (the likelihood that the changes occurred as a result of your treatment, rather than by chance).  They help you predict things, too. If you ever studied anything beyond descriptive statistics in school, you entered the world of inferential statistics.  It’s a scary place for some, but it’s the only place to go if you really want to show causation (that your program really made a difference), and isn’t that what evaluation is all about?

If you need a refresher course on research methods, the Research Menthods Knowledge Base is a great place to start.

The GrantGoddess.com Program Evaluation Resources page has some links to interesting articles on data collection and analysis, as well as a link to two free webinars we have posted on evaluation basics.