IHRIS Collect Design Overview

From IHRIS Wiki
Revision as of 09:00, 18 July 2013 by Sturlington (talk | contribs)

iHRIS Collect is intended as a one-time data collection survey tool built on the I2CE Framework.

Motivation

  • See the following HRIS Blog Entry for a general overview
  • Moving from a paper based system directly to iHRIS Qualify or iHRIS Manage often involves difficulty as the data is not cleaned. So this is really a three step process:
    • collect the data
    • put the data into electronic format
    • clean up the data
  • the data collected may not already be organized into the structure expected by iHRIS Manage. For example, a few places have been interested in using iHRIS Manage. They did not have a Job/Position structure already defined. As a result, they imported/entered a lot of data into iHRIS Manage without a well thought out Job structure.
  • Introduces administrators/developers to the I2CE framework so that they have familiar systems as the move from the data collection process to the routine data entry
  • If data entry clerks are maintained between the one-time data collection process and the routine data collection, they will have a similar system which reduces training. This is the process for the iHRIS Manage implementation in Zanzibar,
  • we have lot of similar functionality and design architecture already built into the system via the custom reporting tool.

Staged Implementation

There are several possibilities for a staged implementation. The stages do not necessarily need to be completed in order. They more reflect a break down into functional units that will ultimately benefit iHRIS Manage, Qualify and Plan.

Stage 1:Branding/Single Form

In this stage, we develop the branding and provided a minimial implementation for iHRIS Collect:

  • Settle on the name (iHRIS Collect)
  • Choose color scheme and update CSS. I vote for beige.
  • Build off of iHRIS Common -- may need to break up iHRIS Common into some sub-modules but probably not.
  • Assume each page to edit only contains one form. Assume the developer can manually define a form in magic data
  • The html template files are built "by hand" by the developer for each form and one which links to all the forms in the survey.
  • A survey page: is a form and its associated template file. Each such page should be packaged as a module and assigned a task to view/edit the data.
  • To enable custom reporting, for each form one needs to define the relationship which contains just the form as a primary form.
  • perhaps we can write a script to automate this once the fields (and their types) for a survey are defined.

Stage 2:Custom Pages

Implements Custom Pages to allow the system administrator to easily create pages based on existing forms in the system.

Stage 3:Custom Forms

Implements Custom Forms to allow the system administrator to easily create custom forms. This can then be used to build custom pages.

Stage 4:Custom Page Displays

As the custom pages will general follow a similar architecture to the custom reporting tools, we can extend the functionality of the custom pages by adding in custom displays. Here are some possible displays:

  • create a PDF version of the survey for paper-based collection. Is this the priority here? Should not be too difficult to do. A few potential pitfalls:
    • what to do with long lists of data, should we display them all and have a check box? or should we leave it as free text and let the data entry people deal with it?
    • what about tiered data lists such as geography (district, region, country)?
  • exporting survey to something be imported to make PDA/hand-held survey tools. also provide need to provide a means to sync the data from the hand-held

Other Documents