Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
The Population vs. The Sample The population Number = N Mean = m Standard deviation = s Cannot afford to measure parameters of the whole population We will likely never know these (population parameters - these are things that we want to know about in the population) 3 Types of Samples • Haphazard sampling – Convenience or self-selection • Quota sampling – Categories and proportions in the population • Probability sampling – Random sampling – Multistage cluster sampling – accuracy (margin of error) & confidence level The Population vs. The Sample The population Number = N Mean = m Standard deviation = s Cannot afford to measure parameters of the whole population So we draw a random sample. We will likely never know these (population parameters - these are things that we want to know about in the population) The Population vs. The Sample The sample Sample size = n Sample mean = x Sample standard deviation = s Cannot afford to measure parameters of the whole population So we draw a random sample. The Population vs. The Sample Does m = x? Probably not. We need to be confident that x does a good job of representing m. The population Number = N Mean = m Standard deviation = s The sample Sample size = n Sample mean = x Sample standard deviation = s Connecting the Population Mean to the Sample Mean How closely does our sample mean resemble the population mean (a “population parameter” in which we are ultimately interested)? Population parameter = sample statistic + random sampling error (or “standard error”) Random sampling error = (variation component) . or “standard error” (sample size component) Use a square-root function of sample size Standard error (OR random sampling error) = Population mean = x+ s . (n-1) The sample Sample size = n Sample mean = x Sample standard deviation = s s . (n-1) The population mean likely falls within some range around the sample mean— plus or minus a standard error or so. To Compute Standard Deviation • Population standard deviation • Sample standard deviation Why Use Squared Deviations? • Why not just use differences? – Student A’s exam scores/(Stock A’s prices): – 94, 86, 94, 86 • Why not just use absolute values? – Student B’s exam scores/(Stock B’s prices): – 97, 84, 91, 88 – Which one is more spread out /unstable /risky /volatile? is the formula for: A.Population standard deviation B.Sample standard deviation C.Standard error D.Random sampling error E.Population mean