Lecture 8 PowerPoint

Lecture 8 PowerPoint - POL 2156B FOUNDATIONS OF RESEARCH...

Info iconThis preview shows pages 1–15. Sign up to view the full content.

View Full Document Right Arrow Icon
POL 2156B FOUNDATIONS OF RESEARCH SURVEY DESIGN
Background image of page 1

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

View Full DocumentRight Arrow Icon
LECTURE PLAN Sampling and measurement error in surveys Types of surveys Response rates: a potential source of error Survey mode: a potential source of error Question wording: a potential source of error Question order: a potential source of error
Background image of page 2
SAMPLING VS. MEASUREMENT ERROR Sampling error is related to selecting respondents for a survey Measurement error occurs when the measurement obtained is not an accurate measurement: the opinion elicited does not accurately reflect the attitude sought Like sampling error, measurement error can be systematic or random
Background image of page 3

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

View Full DocumentRight Arrow Icon
RELIABILITY AND VALIDITY AND SAMPLING AND MEASUREMENT ERROR Random error leads to unreliability: e.g. measurement or sampling techniques will sometimes be “off” because of a certain amount of chance error – what we call random measurement error. Systematic error (or bias) leads to invalidity: measurement or sampling techniques that are systematically “off”
Background image of page 4
SAMPLING ERROR IS THE TIP OF THE ICEBERG -Sampling Error -Non-response bias (entire surveys) -Measurement bias (survey mode and question design) -Random measurement error (survey mode and question design)
Background image of page 5

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

View Full DocumentRight Arrow Icon
TYPES OF SURVEYS Self-administered Mail Internet (probability and non-probability samples) Face-to-face interviews Telephone / Computer-Assisted Telephone
Background image of page 6
RESPONSE RATES
Background image of page 7

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

View Full DocumentRight Arrow Icon
RESPONSE RATES A survey’s response rate is the percentage of individuals selected for participation in a survey (called “respondents”) who actually participate Researchers are concerned about sample-population congruence: how accurately respondents represent the population from which they have been drawn.
Background image of page 8
RESPONSE RATES AND BIAS Survey non-response introduces bias under two conditions: 1) the response rate is low; and 2) respondents and non-respondents are dissimilar
Background image of page 9

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

View Full DocumentRight Arrow Icon
RESPONSE RATES AND BIAS A high response rate has traditionally been considered essential to guarantee that the sample is representative enough to be credible (assumes respondents are more likely to have different characteristics than non-respondents when response rates are low) But studies are beginning to show that lower response rates do not necessarily lead to bias
Background image of page 10
SOURCES OF RESPONSE BIAS Incapacity Noncontact Noncooperation (refusal) Noncontact and noncooperation are the most serious
Background image of page 11

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

View Full DocumentRight Arrow Icon
INCAPACITY The designated respondent is unable to be interviewed because of: ill health language issues (e.g. surveys of new immigrants) illiteracy
Background image of page 12
NON-CONTACT Non-contact: the designated respondent cannot be located, whether because: they are not at home or refuse to answer the phone; a mail questionnaire cannot be delivered or returned; or they do not check email regularly
Background image of page 13

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

View Full DocumentRight Arrow Icon
NON-CONTACT As long as the factors associated with non-contact are random (not associated w/ other characteristics of potential respondents), noncontact is considered an
Background image of page 14
Image of page 15
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 47

Lecture 8 PowerPoint - POL 2156B FOUNDATIONS OF RESEARCH...

This preview shows document pages 1 - 15. Sign up to view the full document.

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