Statistical Methods

or, why Ice Cream causes the murder rate to increase and what this has to do with the Survey


Before we get into my Survey Methods, I should point out to all that I am “mathematically challenged”. Therefore, it is a safe bet for me to state that this is NOT a scientific survey by any means, but is more of a collection of data.

My major in college was Criminal Justice. Toward this goal I took a number of elective courses. One of these classes was in Statistics. On the very first day of class the Instructor stood before us and made this statement: “Statistics never lie, but Statisticians do”. He then proceeded to point out how statistics can be manipulated to produce the results that the statistician wants. The following is the “Ice Cream” example:

Did you know that in New York City the number of murders increase during the summertime? This is a Fact. Ah, but did you also know that during the summertime in New York City the number of Ice Cream sells also increase? This too is a Fact. So, from these two factors what can you conclude? How about: There MUST be a direct correlation between the two. Ice Cream causes the murder rate to increase. So, if you want the murder rate to decrease outlaw the sell of Ice Cream in New York City in the summertime. Makes sense!

Well, it made me NOT want to stay in the class for the whole quarter and instilled in me a great distrust of any and all polls and surveys. Having said all that, let me outline my basic data. Please note that all the statements below are true (to the best of my knowledge) and you can interpret the data as you wish and draw your own conclusions.

The number of paid SCA members (as of 1 April 2000) is approximately: 25,000 The number of those without paid memberships that attend events is: 35% or 8,750 (This figure is based roughly on the number of signed non-member waivers in Meridies.)

So that will give us a “working” number of 33,750 who might attend events and eat feasts throughout the SCA during the running of this survey.

The Feast Survey was distributed through the Internet. There is no way for me to know how many people saw or read the survey. There is no way for me to access any completed surveys unless they were emailed directly to the survey email address.

The number of surveys received was: 689 or just over 2% of 33,750. Now is this enough of a response to make any conclusions? According the many statisticians - Yes. BUT I am not a statistician. I am merely the conveyor of collected data. <smiling> I recommend that if YOU want to know what the people in your area like and/or dislike about feasts YOU conduct your own survey. Remember those little customer survey cards that many restaurants have? Oh, you know the ones...they asked questions like “how was the food?”, “would you eat here again”, “what can we do to make your eating experience better?” Perhaps it would help to have some “after-action reports” of our own.

Now let’s take a few minutes to look at the elements of the survey. I did not request personal background data on the survey, except for Kingdom and length of time in the SCA. I added these two questions, as an after thought, because I “believed” that there might be a great difference based on where people lived and how long they had “played” in the SCA.  But there was little or no difference. (Surprise to me!)

Now many people did not list their Kingdom because it was not on the original survey which was cross-posted. Of those that DID list their Kingdom, I received responses from every Kingdom within the SCA.

I received responses from people who had just attended their first event/feast and from people with more than 25 years in the SCA. Responses came from all levels within the Society: Royals, landed Peers, members of all three Peerages, people with and without titles. I had wondered if the length of time in the Society had a direct bearing on what people wanted at feasts. But, again, there was little or no difference between what someone who was attending their third feast wanted compared with someone who was attending their 100th. (Again, a surprise to me!)

The closest thing to an “actual statistical survey” was the one week survey that was conducted on the Meridian List. It was designed with multiple choice questions to produce actual percentages. Although this survey was helpful with insuring that the data was uncorrupted, it did not allow for the personal comments that I have found so valuable. (See Feast Survey with comments by submitters)

For those individuals that are more “mathematical” (or just curious) you may be interested in the detailed numbers from the Meridian Survey.

Although numbers have their place in the scheme of things, it was the personal comments that I found truly revealing. If I had one suggestion to make I would recommend that you read over the many comments that people freely shared with me. Comments that enlightened me to other ways of doing things. Of other viewpoints and perspectives. Of ideas and concepts I had not yet considered.... even after 25 years (plus) of “playing the game.”   For this is the REAL reason for the Feast Survey - it is a starting place for the interchange of ideas.
 
 
<-- The Original Survey (Previous section) The Numbers from Meridies (Next section) -->
Back to the Feast Survey main page All text © 2000 Raine Phillips

Visit my CafePress store!