I am making part of group that will collect data all over the country from a survey. There will be 3000 thousand HIV patients included on the sample.
There are one big form with 10 pages and there is another for lab data. The study will take 6 months.
The questions:
Can I use DHIS for this? Can you provide recommendantions.
The objective is to reduce time on database design and I could spread this much more quickly and the madness of dealing with users DHIS would already do for me.
If DHIS can be used how I do the pages separation?
IMHO, DHIS2 is not well suited for survey data. It depends on the nature of the survey really, but remember that you must know the nature of the organisational hierarchy (or beneficiaries in this case) before hand. Most surveys also have some type of skip-logic/validation which DHIS2 does not have at the moment.
We considered using DHIS2 here in Zambia for a survey, but it sort of felt like trying to squeeze a square peg in a round hole. We ended up using PDAs for the primary data collection, which was then processed and cleaned in statistical software, and eventually, the aggregate data was transformed and imported into DHIS2.
You may want to consider giving http://limesurvey.org/ a try. It is a very powerful tool, and has many more features which are suited to surveys than DHIS2.
For paper based surveys, CSPro is a good choice.
There are lots of PDA based mobile data collection tools as well.
At the end of the day, it really depends on the nature of the survey. If you have a known orgunit hierarchy (or I guess in this case, a list of beneficiaries) it might work for your purposes, and their are obvious advantages to having the data directly in DHIS2, but you may want to consider other software (such as LimeSurvey, CSPro) for primary data collection, followed by statistical analysis and cleaning, and finally, export into the aggregate system.
Just my two cents.
Jason
···
On Thu, Nov 3, 2011 at 8:30 AM, Orvalho Augusto orvaquim@gmail.com wrote:
I am making part of group that will collect data all over the country from a survey. There will be 3000 thousand HIV patients included on the sample.
There are one big form with 10 pages and there is another for lab data. The study will take 6 months.
The questions:
Can I use DHIS for this? Can you provide recommendantions.
The objective is to reduce time on database design and I could spread this much more quickly and the madness of dealing with users DHIS would already do for me.
If DHIS can be used how I do the pages separation?
IMHO, DHIS2 is not well suited for survey data. It depends on the nature of the survey really, but remember that you must know the nature of the organisational hierarchy (or beneficiaries in this case) before hand. Most surveys also have some type of skip-logic/validation which DHIS2 does not have at the moment.
We considered using DHIS2 here in Zambia for a survey, but it sort of felt like trying to squeeze a square peg in a round hole. We ended up using PDAs for the primary data collection, which was then processed and cleaned in statistical software, and eventually, the aggregate data was transformed and imported into DHIS2.
You may want to consider giving http://limesurvey.org/ a try. It is a very powerful tool, and has many more features which are suited to surveys than DHIS2.
For paper based surveys, CSPro is a good choice.
There are lots of PDA based mobile data collection tools as well.
At the end of the day, it really depends on the nature of the survey. If you have a known orgunit hierarchy (or I guess in this case, a list of beneficiaries) it might work for your purposes, and their are obvious advantages to having the data directly in DHIS2, but you may want to consider other software (such as LimeSurvey, CSPro) for primary data collection, followed by statistical analysis and cleaning, and finally, export into the aggregate system.
Just my two cents.
Jason
On Thu, Nov 3, 2011 at 8:30 AM, Orvalho Augusto orvaquim@gmail.com wrote:
I am making part of group that will collect data all over the country from a survey. There will be 3000 thousand HIV patients included on the sample.
There are one big form with 10 pages and there is another for lab data. The study will take 6 months.
The questions:
Can I use DHIS for this? Can you provide recommendantions.
The objective is to reduce time on database design and I could spread this much more quickly and the madness of dealing with users DHIS would already do for me.
If DHIS can be used how I do the pages separation?