Class Notes (807,255)
Canada (492,677)
Sociology (3,130)
Lecture 4

Sociology 2206A/B LECTURE 4.docx

6 Pages
Unlock Document

Western University
Sociology 2206A/B
William Marshall

2206 Monday, January 30, 2012 Measurement What is it? -The process of creating measurable concrete variables from abstract concepts -Extend the senses (empirical-much less sensitive to change than senses; yields more consistent between findings ex: thermometer =more accurate temp than someone just touching a pole and gauging that way) -Tends to be more of a concern for quantitative researchers Measurement Differences: Quantitative vs. Qualitative -Timing -Type of data collected -Linking concepts to data (pre-planned vs. reflective/interactive) The Measurement Process -Quantitative=deductive -Qualitative=inductive -Both involve conceptualization (the process where you think through the meaning of the construct) and operationalization (when you take the conceptual def and turn into different measurement techniques) Steps -Identify concept of interest -Develop a conceptual definition Decide on units of analysis -Operationalize to create a variable Links concept to measurement technique Process of Quantitative Operationalization ** see pg 107 in textbook for “Abstract Construct to Concrete Measure” diagram Process of Qualitative Operationalization Empirical Observations Working Ideas Concepts Generalizations/Theories (**Not every qualitative researcher will use this process—more flexibility) Reliability and Validity -Reliability: the consistency or dependability of a meansure Does the measure consistently give the same results? -Validity: Truthfulness of a measure Is it measuring what the researcher thinks it is measuring? (**No perfect validity of reliability—they are ideals) 2206 Monday, January 30, 2012 Types of Reliability -Repeatability/Test-Retest  results should be similar after re-test at a later time -Concordance/Inter-Rater if two different researchers test the same thing they should both end up with he same or very similar results -Equivalent forms/ Parallel Forms When 2 different tests are measuring the same attributes; -Split-Half Asking similar questions on two halves of a survey--checking to see if responses changes as the survey progresses; check for lying; checking how consistent people are being Improving Reliability -Clear conceptualization -Use of most precise (or highest) level of measurement possible EX. Instead of asking if you are young or old; ask date of birth -Use of multiple indicators -Use of pretests, pilot studies, replication Validity -A test is valid when it measures what it’s supposed to -If a test is reliable, it yields consistent results -Relationship between Reliability and Validity? You cant have validity without reliability; having a reliable measure doesn’t guarantee you have a valid measure As one gets better, the other one usually gets worse—need to find a balance of the two Relationship Between Reliability and Validity ** See “A Bull’s-Eye=A perfect Measure” Diagram in Textbook Types of Validity -Face Validity An expert saying this is true -Content Validity Special type of face validity—all the important parts of the conceptual definition are included in the measure -Criterion validity Comparing to something already proven valid 1. Concurrent validity: less rigorous; indicator associated with one already proven in another (EX. People who score well in the LSATs will probably do well in Law school) 2. 2206 Monday, January 30, 2012 Internal and External Validity -Internal Validity Can alternative explanations for change in the dependant variable be eliminated? -External Validity Generalizability Extraneous Variables can Interfere with internal Validity: 1. History (did something happen out of your control during study that could have impacted the dependent variable?) 2. Maturation (normal developmental processes happening throughout the duration of your study; ie. Studying youth—they get older) 3. Testing (do the pre-test scores effect the post-test scores? Have they already seen survey or practiced skill since last survey?) 4. Instrumentation (did you change the way you measure the dependant variable from time 1 to time 2?) 5. Statistical regression (the extremes moving more towards the middle; observe biggest change in the most extreme group—ex. Poorest readers will always improve more than advanced readers) 6. Selection (want groups to be as close together from the start) 7. Experimental morality (people move, people die, people decide they don’t want to participate anymore—(longitudinal study) initial study to have as many people as possible to account for losing people) 8. Selection-maturation interaction (two groups were diff at beginning of research but normal developmental changes occurred for one group but not for the other –maturation happening for one group but not for the other) Other Research Effects -Experimental Bias If our expectations are communicated to participants can change the outcome, and change course of the larger experiment; participants will want to make you happy, so they will do things to try and give you the results that you want) -Placebo Effect  Receiving no treatment, but are getting better -Hawthorne Effect People will change the way they act if they kno
More Less

Related notes for Sociology 2206A/B

Log In


Don't have an account?

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.