Zaldy Rodriguez

 Background 7/16/13 – Zaldy discussed the current state process for metrics and possibilities for the future. We assumed that there will be no change to the current paper process for survey collection. We also assumed that the data will eventually be entered into a relational database. The scope of this document will look at the encoding process inbetween survey collection and the database. Considerations: 1. Limited accessibility to internet in the bases 2. The current Dropbox method produces conflicted files when multiple people are using the same file at the same time Option 1: Mail surveys to Bacolod for data entry All encoders will be hired in Bacolod. Pros Cons Internet access in Bacolod is reliable Significant changes to the job of Metrics leads for each base (proposed change below under recommendation) Flexibility on encoder hiring – can be more precise about how many to hire and easily transition to perminant employees or not as desired Need to trust the mail service or we will lose the surveys in route to Bacolod.* Change management will be easier since all the encoders are in one location. Easier to train, hire, and manage with scale. Time delay for surveys to get entered. Scaling is easier for base leadership as fewer tasks are assigned for each base to do independently. Scaling is harder at the national level since the would need significant space and staff for encoding. Surveys could be easily scanned from one location if the decision is made to switch to a digitial archive *expense of mail is not a con since surveys are mailed to Bacolod for archiving anyway Option 2: Scan surveys at the bases All encoders will be hired in Bacolod. Paper surveys can destroyed after they are scanned in the bases. Scanning software will automatically send the scanned files from the printer to a Dropbox (or BTsync) folder. Pros Cons Eliminates mailing the surveys – removing expense and any risk of lost surveys Cost of setting up scanning – both in terms of the scanners and time required to set up adequate reliability in each base. Encoder hiring and change management pros from above still apply Conflict files may still occur (though should be far less common with image files than with spreadsheets) Easier archiving for staff in Bacolod since they never need to process a paper survey Encoders work from images instead of paper surveys – may need 2 monitors each. Uncertainty about what this will do to productivity Same Metrics leads con from above still applies Option 3: All bases will do their own data entry Each base will have access to database data entry tools. Pros Cons No staffing changes are required Issue Management will be harder since the people most familiar with the software will not be physically present. Training of encoders will have to be done by Metrics leads at each base, making it harder to share improved techniques across ICM. Recommendation After discussing the pros and cons above, Zaldy, Lincoln, and Nate recommend option #1. Mail has proven reliable. All non-simple steps of the process are also located in one place which will be a big advantage for maintaining the process and any future change management. The one significant change this necessitates is the job description of the metrics lead in each base. The metrics lead continue to manage the surveys teams exactly as they do now. However, since they will no longer have local encoders to manager, we propose that the metrics lead pre-audit data entry by personally encoding 5-10% of the surveys from their base before mailing them to Bacolod. Encoders in Bacolod will not know which surveys have been pre-audited. The database will track differences and auto-email summary statistics as desired. Since we are implementing a new encoding system starting in September, this is the ideal time to make any changes to the base org structure. We should not engage in the effort to train all bases on the new data entry process if they will not have the responsibility to carry out that process long-term. Next Steps: August 16 Helen and Dan make a decision on the options above August 19-21 Zaldy, Lincoln, and Nate will be in GenSan to do a test run of survey data entry September Batch 2 Pre-survey data entry (new process fully implemented) September 24 Nate returns the US – will be available for some report support, but not physically present for the remaining data entry October Batch 1 Post-survey data entry Supporting document: ICM Metrics Data Process.ppx has workflow diagrams and technical details