Home > Statistics Every Writer Should Know > The Stats Board > Discusssion
Variance formula
Message posted by Sinzana (via 216.104.6.40) on August 3, 2001 at 4:38 PM (ET)
What is the basic formula to figure out a variance between a series of numbers?
READERS RESPOND:
(In chronological order. Most recent at the bottom.)
Re: Variance formula
Message posted by JG (via 128.8.22.17) on August 4, 2001 at 5:26 AM (ET)
Do you mean variance as defined in most statistics books or just some measure of variation. There are many measures of variation besides variance and standard deviation.
Re: Variance formula
Message posted by sinzana (via 152.163.197.178) on August 4, 2001 at 8:53 PM (ET)
The reason for the question is that I am looking for a variance / a percentage from an average. I need to calculate what the variance is from the average on the largest and the smallest number that is included in the average. Example: I have a series of eight numbers. The numbers are 25, 36, 27, 48, 59, 40, 51. The average is 35.75. I need to figure what the variance percentage is of the largest number (59) from the average and the variance percentage of the smallest number (25) from the average. The need for this is that I am going to allow a process to continue, as long as my next number, which I will not know what it is ( the process is automated), does not exceed the maximum allowable variance that I have just calculated. Do you have any ideas for this type of calculation?
Re: Variance formula
Message posted by JG (via 128.8.22.227) on August 6, 2001 at 8:10 PM (ET)
Look at control charts. This usually involves the assumption that your data is normally distributed. Also, look at box-and-whisker plots.
Re: Variance formula
Message posted by RP (via 131.247.174.22) on August 8, 2001 at 12:43 PM (ET)
I too have a variance question. I am studying 2 groups of people. 1 group gets 13 pieces of information in a row before providing an assessment. The other group receives the same 13 pieces of information, but are asked to make an assessment after each piece. I want to show that the latter group is becoming more homogenous at the 13th piece of information than the former group. However, it is intuitive that 1 group starts at nothing will have a wider variation than the other group who is receiving information all along. Therefore, I am doubling the amount of information (i.e., 26 pieces). Thus, my question is how can I measure variation for both groups? Again, I want to show that the variation or the group making continuous assessments is lower than the other group at the 13th and 26th piece of information (those are also the points in time the other group is going to make their first 2 assessments). Thank you.
Re: Variance formula
Message posted by JG (via 128.8.22.67) on August 8, 2001 at 6:53 PM (ET)
It is not clear to me what you are trying to do. Also, are these people aware of each others assessmets. However, the following topics have been studied by others and may be useful to you. They are - sequential decision making, developing a consensus in expert opinions by means of controlled and partial feedback, group decision making, etc.
Re: Variance formula
Message posted by RP (via 131.247.167.78) on August 8, 2001 at 10:28 PM (ET)
Thank you for the tip. No, the groups don't know about each other. Also, I have looked at the research in your tips, especially sequential decision making. Unfortunately, only 1 of the 2 groups is receiving sequential information with the other group receiving information more in a lump sum. I guess then my question is similar to Sinzana's earlier question, can you think of any variance formula(s) that would be appropriate to study variation between my 2 groups?
Your $5 contribution helps cover part the $500 annual cost of keeping this site online.
