Summary of comments to Salil's question

In response to Formats of Rounds - I, Salil had a question to ask on awarding split points in an answer to more than 1 team.
Principal comments:

Arnav:

The quizmaster should in general follow a policy of not saying if a particular answer given by a team is completely wrong or partially right, and should move on. In the eventuality that another team gets the complete answer, they should be given all the points, otherwise the q/m can come back to any team that had answered partially correctly to award partial points. Mine:

Incidentally, this point took centrestage during VIT's quiz where it was felt that the hosts were trying to please everyone by awarding half points to many teams (once overshooting the value of 10 per qn), or were hesitant in deciding whom to award points to etc.. This issue crops up principally in responses to Connects as the routine observation is that no team manages to answer the question to full satisfaction. Usually they get the relation or the central theme correct and can explain only part of the connection between the elements and the theme - they then win half points for their effort. A team ahead in the passing order then completes the relation and takes the remaining points. Sometimes, it even takes more than 2 to complete the entire jigsaw. This causes even more headaches, for it seems inelegant to award fractional points or more than the original value of the question.

In my opinion, the following can be done by the question setter. First, while creating the question, try and understand whether there will be a possibility of a fractured answer. (In the case of a connect, this is highly likely. Also, in the example quoted by Salil, there are three names required in the answer - another candidate.) If this possibility exists, perhaps the question can be modified to expect only one clear cut answer? (In Salil's e.g., is there a need for all three blanks to be filled, esp. given that IG & SG will be the obvious guesses? Can only the relatively lesser known Bansi Lal be the hidden one?) If so, the problem is averted. If (as in the case of connects) this cannot be avoided, then one needs to think what is the most important aspect of the answer. Is it requisite for all elements to be explained, or is it fine for the theme (the "funda" as we call it) to be revealed? For instance, in the "Elements" connect that I have mentioned here, I have no right to expect an explanation for each item! In Chakravyuuh IV as well, the connect for the Space Missions would be resolved with the theme, instead of the organisers insisting on each individual explanation. In the event that both the theme and each explanation is essential (connects with 3 or 4 items involved), then it must be made clear to the teams how the scoring will be done, so that there is no grumbling later. I faced this potential issue in a round I had done as part of my PULSE quiz, where I had a Visual Connects round called Split Personality in which a set of 3 to 4 images connected a person. In this case, the name of the person fetched 5 while the explanation (of a majority of the items, not all) fetched the remaining 5. The round passed off peacefully ;-)

This problem has no quick and simple solution for all cases - I think the questions can be finetuned to avoid the problem in most cases. Also, the person asking the question needs to be clear in his/her mind and consistent in implementation. The problem is compounded when more than one person is conducting the quiz. A decision prior to the question needs to be arrived at; even perhaps to the extent of a points breakup for each such question agreed upon in advance. Perhaps a radical method can be followed only for these types: get the teams to write down their own explanation for the connect, inclusive of the main connection and each item's relation to it (the writing is to prevent wholesale advantages to the teams further down the order). Then they read out in turns, and the first in the passing order to provide the best (if not complete) gets points. This can avoid cases where a team later in the order needs to fill in only one missing piece, having used the previous answer to their own advantage - even if this team would have got it otherwise is hard to prove. I don't know if this can work practically or if it is an ugly hack but a workaround like this may be important in a quiz where there is a profusion of such questions.