Survey Data Analysis

August 28, 2006

I came up with my own idea for a project to do at work again this term. This time I noticed that the surveys that were being used on our department's website could use a little work. With that in mind I thought of how I could fix it and brought my ideas to my boss. Shortly after that I got the ok and was on my way implementing my ideas.

The way the survey results were being analyzed before was terrible. All of the results were being stored in an excel spreadsheet that was ridiculously difficult to interpret anything let alone find something specific to look at. So my first task was to recode the actual surveys using PHP to interact with a MySQL database to store the results. With the results being stored in a database, analyzing the data had numerous possibilites.

I implemented an AJAX application to view the results of the surveys. The application includes many unique features that make it stand out, including NiftyCube and DLCalendar. The results from each question are displayed in a table as text followed by a corresponding graphical representation. The graphs are created using a combination of PHP and CSS to calculate and display the information. Another unique feature of the graphs is that each bar is colour-coded with respect to percentage.

The options panel includes a drop down menu to select a survey, another drop down menu to select a question corresponding to the survey selected, and an option to view results in a specified date range. The date range option is where the DLCalendar comes into play, allowing for an easy selection tool to change the date. The option panel also features a toggle button to allow it to hide, generating a larger viewing area for the data results.

The AJAX comes into effect whenever an option is changed within the options panel. Any change made sends out an XMLHttpRequest allowing the data to be updated without the page reloading. After optimizing the code, any changes made looks almost instant and thus resulting in a fast, clean, user-friendly application.

Jason Smith

Reader Comments

  1. Craig says:
    April 13, 2007 8:52 am

    Cool... but where do I find this? A link would be nice! I've wanted to ask this since you posted this article.

  2. Jason says:
    April 17, 2007 2:29 pm

    I was going to implement a mock up version on my own site with some random data but you can just take a look at the real thing:

Comment On This Article

All HTML tags will be removed, but URLs will be automatically linked. Quality over quantity – comments will be deleted if they do not add to the discussion in a substantial manner.