[QUOTE=FireWorks;332895]The last options are “3+ times” which means “3 or more time”. Please stop trollin[/QUOTE]Having worked extensively and professionally with gathering input from people using surveys, I can assure you that he is not trolling.
You may only want empirical numbers-based responses, but participants simply neglect to complete surveys if they feel that they are inadequately designed and don’t give them the proper outlets to field their responses. This is called Response Rate, and you certainly didn’t get a vote from me.
For one thing, Survey scales need to be balanced, or their Response Rate plummets.
Furthermore, you aren’t really interested in the Accumulated number of Wins or Losses that players have seen, because this varies based upon extent of play experience (i.e. Having personally accumulated hundreds of games, so what if I’ve seen them Lose 3+ times…). What you are interested in, is the Frequency or Proportion of Wins and Losses that players have witnessed.
The fact is that rates people will report are not objective, and you will have a difficult time quantifying them, because people don’t literally know what percentage of the time they will Win or Lose on Container City.
They can tell you the gist, however.
A common Likert scale for this purpose is:
- Always
- Often
- Sometimes
- Seldom
- Never
If you want to further anchor those values, you can try to do so with percentages or “out of ten” anchors.
i.e.
- Always (100% of the time)
- Often (~75% of the time)
- Sometimes (50% of the time)
- Seldom (~25% of the time)
- Never (0% of the time)
and
- Always (10/10 times)
- Often (7-8/10 times)
- Sometimes (5/10 times)
- Seldom (2-3/10 times)
- Never (0/10 times)
Finally, after all is said and done, and you’ve put together your survey, you can never assume that it is functioning as you’ve intended. You need to look at the data, and examine whether it is skewed or not. The fact is that people’s interpretations of language vary, and people are also susceptible to what psychologists refer to as Biases.
Assuming that you are going for something worth reporting, you need to examine how it compares to valid reference results (i.e. psychology surveys) or objective results (i.e. the actual statistical frequency of Resistance full-holds on the last objective of Container City).
Even if a survey isn’t about being formally reported, you need to be prudent about drawing conclusions based upon trends that you observe in your survey, given the limitations of your method and sample size, etc.
At the end of the day, a survey is only viable if it is used as a tool for action.
Too often, surveys are poorly designed and are little more than GIGO-machines (Garbage In; Garbage Out) because they clearly have an agenda they are either trying to push (or confronting a reality that the reviewers and designers are too scared to face). Guess which category your poll falls into…
EDIT:
For what it’s worth, I’d say that Resistance seldom accomplishes a full-hold on the final objective of Container City.