El Cuestionario

El Cuestionario (formerly known as Rate Yourself) is a tiny web application to display and evaluate single-page questionnaires.

A questionnaire (questions, their answers, and score ratings) is defined as JSON in a single file.

elcuestionario_questionnaire_preview.png

Screenshot of a partially completed questionnaire.

The questions are presented in the order they are defined in the data file while the answers are shown in random order.

Once all questions are answered, the user's score is calculated based on the answers' weight (as defined) and is presented with a suitable comment.

elcuestionario_result_preview.png

Requirements

Installation

Install dependencies:

$ pip install -r requirements.txt

Make the script to run the application executable:

$ chmod +x runserver.py

This script (which uses the built-in web server) is fine to get up and running quickly. However, to actually serve the application on the Internet, please consider the other deployment options described in Flask's excellent documentation.

Configuration

Take a look at data/example.json regarding how a questionnaire is defined. Just copy the file, adjust its content and update the value of the SURVEY_FILENAME variable in the main script accordingly.

Title

A questionnaire has a title:

"title": "A Bunch of Questions",

Questions and Answers

Each question should have multiple answers.

Each answer must have weighting, defined as a float number between 0.0 and 1.0.

Exception: I sometimes use a value of 1.1 for "bonus answers" which allows for an overall score of more than 100%.

To calculate the overall score, the weighting of each question's answer will be used.

Example:

"questions": [
  {
    "text": "What's your favorite color?",
    "answers": [
      { "text": "blue",      "weighting": 0.7  },
      { "text": "green",     "weighting": 0.5  },
      { "text": "yellow",    "weighting": 0.1  },
      { "text": "red",       "weighting": 0.25 },
      { "text": "none",      "weighting": 1.0  },
      { "text": "checkered", "weighting": 1.1  }
    ]
  }
]

Ratings

Ratings can be defined to add a text to the result based on the achieved overall score. Each is bound to a score range.

The minimum_score value sets the threshold for each rating. The adequate rating will be chosen by finding the one with the highest minimum_score value that is lower than the score. For example, with a score of 53 % and ratings with minimum_score values of [10, 20, ..., 90, 100], the selected rating will be the one with a minimum_score value of 50 since it is lower than 53 and the next higher minimum_score value, 60, is not lower than the score of 53.

Therefore, a rating's minimum_score value defines the minimum score one has to gain to be given that rating, as long as no other rating is more suitable considering its minimum_score minimum.

Example:

"rating_levels": [
  { "minimum_score":  0, "text": "OMG, please waste time with something else!" },
  { "minimum_score": 40, "text": "Not bad." },
  { "minimum_score": 70, "text": "Looking good." },
  { "minimum_score": 90, "text": "Yeah, great result!" }
]

Rating levels are optional. If none are defined in the data file, to rating text is shown on the result page.

Usage

Start the application with the example configuration:

$ ./runserver.py data/example.json

It will spawn a web server on port 5000.

To access the questionnaire, point a web browser to http://localhost:5000/.

You can also specify a custom port to listen on as well as enable debug mode:

$ ./runserver.py --debug --port 8080 data/example.json

This will make the questionnaire available on http://localhost:8080/ and provide an in-browser debugger in case an exception is raised by the application.

Changes

First released on 2006-04-26.

Notable changes since the first release:

  • The data format for a questionnaire changed from XML to JSON.

  • WSGI (via Werkzeug) has replaced CGI as the interface to the web server to support more deployment options.

  • Jinja has replaced Kid as the template engine.

  • The original script evolved into a Flask application with separate modules and templates.

  • Tests have been added.

  • Naming has been adjusted to follow PEP 8 more closely.

  • Added support for Python 3. Dropped support for Python 2.