review sample

review sample - SELECTINGASAMPLE...

Info iconThis preview shows pages 1–11. Sign up to view the full content.

View Full Document Right Arrow Icon
SELECTING A SAMPLE  Copyright 2010 Margaret DeFleur
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
SELECTING A SAMPLE Imagine we want to measure the IQ of each person who entered a graduate program in the U.S. in the past 5 years, to assess the relationship between IQ and completion of degree program. Population = 10,000 1. not possible to measure all 2. you decide to sample from the population 2 MC3018 Foundations of Media
Background image of page 2
SELECTING A SAMPLE Using a sample as an estimate of the population: 1. You want to generalize the results of the sample to the population 2. You want to say with confidence that the sample results reflect what we would find if we measured everyone in the population 3 MC3018 Foundations of Media
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
SELECTING A SAMPLE Important Sampling Issues: 1. How large must the sample be? Does the size of the sample make a difference? 2. Sampling Error (the variability present in any sample) 3. Techniques for sampling 4 MC3018 Foundations of Media
Background image of page 4
SAMPLE SIZE IQ has a range of values (avg. for population = 100) population of graduate students = ? If we select sample of 10 students, IQ values will be scattered, not just near population mean Better results with sample of 100 With sample of 1,000, chances are very good that sample mean we find will be same as that which exists in our population 5 MC3018 Foundations of Media
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
CURVE OF DIMINISHING RETURNS As sample gets larger, curve of diminishing returns operates: 1. very little accuracy with sample of 10 2. accuracy increased with sample of 100 3. sample of 1,000 really accurate 4. sample of 5,000 gives only slight increase in 6 MC3018 Foundations of Media
Background image of page 6
CURVE OF DIMINISHING RETURNS 7 MC3018 Foundations of Media
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
CURVE OF DIMINISHING RETURNS MC3018 Foundations of Media 8 Considerations: 1. Ideally, we want sample mean = population mean; accuracy of estimate is perfect 2. Must consider overall goal (to generalize to population?) as well as technique for sampling
Background image of page 8
WHAT IS THE IDEAL SAMPLE SIZE? MC3018 Foundations of Media 9 There is no simple answer to this question: 1. You want the maximum size for the resources you have 2. Beyond 1200-1500 gain in accuracy is slight 3. How many can you afford to study? Cost is a factor 4. Also depends on the importance of the results http://www.raosoft.com/samplesize.html
Background image of page 9

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
SAMPLING ERROR MC3018 Foundations of Media 10 Sampling Error:
Background image of page 10
Image of page 11
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 39

review sample - SELECTINGASAMPLE...

This preview shows document pages 1 - 11. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online