Summer Milestone 4 Micro+Macro : Analysis and Addressing Deeper Questions
Contributors to this Wiki
Kajal Gupta: @kajalgupta
Surabhi Ingle: @surabhi_i
First Micro+Macro discussion : Analysis and Addressing "Levels"
Levels are an important measure of every worker's professional expertise.Hence, the reputation system must be a fair one, depending on not only one criteria but various quantitative as well as qualitative aspects. Hence, workers on the platform must have certain levels attached to their profile.
Goals of levels:
-Fair judging of the workers by the Requesters
-Increasing of trust between the two parties
-Fair opportunity to new workers on the platform
-Mapping requester to right worker .
-Giving more working opportunity to worker
Any platform consists of two categories of workers:
a. The old workers who have completed atleast one task
b. The new workers who have not completed any tasks.
The system of level allotment needs to be different for both these categories. Hence, We may want to propose few things here:
-New workers need to get themselves graded or levelled. There are various ways they can do so.(see the image)[]
-The worker has the flexibility of making the method by which he was allotted those levels. eg Level 5 Python programmer (Self-assessment method). However, displaying of the method will help the requesters select workers with better efficiency
-The worker has the flexibility of choosing any of the methods. He can also start with no levels tagged to his profile in case of micro tasks like filling up of surveys. The default level will be level 1.
-In case the worker feels that he has been misjudged, he can re-appear for the above mentioned methods. This is only applicable once per level up-gradation.
-The best off both the results will be selected as the final level.
Methods of level allotments
1) Certification and Tests: This is a paid method of level allotment. Here, workers pay for taking the certification tests. The funds raised by the workers will be distributed amongst the set of other, high level workers who evaluate this work for a fixed pay. The test result will decide the level of the worker. The work done during the test can be made public by the worker so that his work can be better evaluated by the requester in person._iIn case the worker wants to add levels from another platform he has been working on, he can skip this step by uploading his level from the other platform.
2) Volunteer for social activity: The workers can work for helping communities working for social causes, eg, designing a logo for a school of differently abled children at a minimal amount of pay. This level is exclusively for the old workers, the reason being that, we cannot guarantee what level of skill a new worker has, hence, we cannot guarantee of quality to these clients. Old workers will work well on these tasks, since they guarantee them a level and will be able to do the job above a certain standard of quality.
3) Involving a mentor/requester/high profile worker:The worker can fill in the skill-set he claims to possess. The platform allots them mentors. These mentors can be a requester or a high-levelled worker. These mentors have to review their work i.e. allot them levels. The mentors can evaluate using any methods, like, tests designed by them, interviews etc. This method can be paid or unpaid, depending on the mentor. Reputation badges or any other incentive may be given to the mentors.
4) Linking of profiles from other crowd sourcing platform : In case the worker wishes to add his profile from any crowd-sourcing platform he has previously worked on and/or is currently working, he can do so by choosing this option.
5) Quasi-qualitative method: This method is exclusively for old workers. The workers can have their levels upgraded by qualitative feedback like level of quality of work produced, etc and also by quantitative methods like time taken for completion of tasks, etc.
Our proposal has been framed with keeping the discussion of first Micro+Macro meeting in mind and as an attempt of minimizing the chances of bad end result and increasing the fair level-fair work-fair pay policy.Also, the interface consists of an ”Ask for help” button, which will help a worker in deciding which method to go for in case he is confused.Our scale both for Micro as well as Macro tasks.
Some of the challenges our proposal might face are:
-Choosing the right mentor/requester/worker as a mentor.
-Type of incentive to be given to them.
-Mapping the worker to right requester.