Flashlight Online: Advanced Topics

by Larry MacPhee, e-Learning, NAU

The URL for Flashlight is http://ctlsilhouette.ctlt.wsu.edu and online help help icon is available even without a login name and password.

Flashlight Basics | Common Problems | Flashlight Advanced


Now that you've built a simple survey with the flashlight tool and learned how to navigate within the web based interface, you'll want to know how to create more sophisticated surveys that better meet your needs.

Advanced Topics:

1) How Do I Copy a Survey? The "copy" copy icon button allows you to create a duplicate copy of your survey.  This is helpful if you want to use the same survey at different times (and keep the results separate), or if you started a survey and then decide you want to add or modify some questions. (Remember that the copy is made from what is stored on the remote server and not from what you may be editing in your browser. Always save your edits by clicking OK or Submit before you make a copy.)

2) How do I ensure that each person can only take the survey once? On the "survey properties" survey properties icon page, you can use "validation" to ensure that each person only takes the survey once. This involves assigning a unique ID (or an ID and password) to each respondent. Anonymous surveys in Flashlight allow multiple submissions by the same person. If your survey must be both exclusive and anonymous, the best you can do is to have a 3rd party assign user IDs and not report who is who to the person doing the analysis of survey results. (In online survey design, there is an inherent trade-off between security and anonymity. There are many technical tricks a survey designer could use, such as setting cookies and logging IP addresses, to prevent users from taking the survey more than once, but each can be defeated by a sophisticated user, so the only sure solution is to assign each respondent a unique ID.)

3) How do I set up user IDs for survey participants? You can enter IDs manually or paste them in (from Microsoft Excel for example) if you need to enter lots of user IDs (or IDs and Passwords). To manually enter user IDs and passwords, first set the survey to require user IDs and Passwords (on the "survey properties" page) and then click the "respondent ids" button. Enter user ID and password with a comma between them and only one user ID and password per line as shown:

4) How can Excel help me set up user IDs and Passwords? Download this Excel Spreadsheet and copy and paste the contents of Column E into the box in the "Add New Respondent IDs" window. This assigns a random, whole number password between 1 and 1000 to users who are sequentially named "person1", "person2", etc. If you need more names and passwords, use Excel's copy and fill down commands. Notes: the Excel spreadsheet uses the RANDBETWEEN function that is part of the optional "Analysis Toolpack" installation. To check to see if you have it installed, go to Tools/Add-Ins... and see if Analysis Toolpack is listed. If not, you would need to install this add-in from your Excel installation CD. Also please note that each time you open the Excel spreadsheet, different random numbers will be generated, so be sure to paste them into Flashlight all at once and make a printed copy for your records!

4) What are "categorical variables" and when should I use them? You can give the same survey to different groups of people and categorize the responses by group for later comparison. For example, you could use a generic course evaluation survey and have the ability to look for differences in student satisfaction over several semesters and with multiple instructors. You can have up to nine variables. Click the "categorical variables" button at the top of the window and follow the directions on screen. Set up the variables, such as course name, semester, and instructor name, and set the values for each (the courses, semesters and instructors you would like to compare). When you are done, Flashlight will generate a unique URL for each combination (for example, ENG_100, taught in Summer 04 by Jones) and you will give each URL out only to that group.

5) How do I export the data to analyze in Excel or SPSS? Click on the "analyze" button and select "Follow this link to download data". Select the survey by name and follow the directions on screen. Flashlight will export a CSV (comma separated values) file that Excel or SPSS can import.

6) How do I use templates? Why design your own survey when there are lots of pre-made surveys to choose from and customize for your needs? First, create a "new survey" (you must start with a survey that has no questions in it) and then choose the "templates" button. A list of templates will be displayed. After you apply a template, you cannot undo it so be sure to click on "view template" to preview it before you proceed. Edit the text of the questions, de-select ones that you don't need, and re-order the questions as desired with the up and down arrows. Click the "OK" button when you're done.

7) How do I edit the HTML on a survey to customize the appearance? The "Customize HTML" customize html icon command enables you to change the look and feel of your survey. Warning: Once you begin customizing a survey, you can no longer add a question without losing the customization you've done to that point. When creating a survey, you're actually doing two things at once: creating the look of the survey (which you can modify using html) and creating the database that receives and analyzes the data. Starting the customizing process also ends the ability to modify the database. Customizing, however, does allow many kinds of changes, including:

* Be cautious about using an HTML editor to change the order of questions however. It's quite easy to make changes that inadvertently cripple the survey. FrontPage seems to be particularly unpredictable as a tool; our users have had better luck with Dreamweaver. As awkward as Flashlight Online's tools for reordering questions are, it's usually best to use them to reorder questions and only risk using html editing for this purpose if absolutely necessary and after making a backup copy of your survey.