Measurement Notes Soc 3163
Arkansas Tech University
Popular in Intro to Sociological Research
verified elite notetaker
Popular in Sociology
This 4 page Class Notes was uploaded by Krista Lindenberg on Friday February 19, 2016. The Class Notes belongs to Soc 3163 at Arkansas Tech University taught by Dr. Mikels-Schlutterman in Spring 2016. Since its upload, it has received 16 views. For similar materials see Intro to Sociological Research in Sociology at Arkansas Tech University.
Reviews for Measurement Notes
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 02/19/16
Tuesday, February 16, 2016 Measurement Issue Operationalization – process of connecting concepts to observations When we conceptualize, we specify what we mean When we operationalize, we identify specific observations that we will take to indicate that concept in empirical reality Options for Operationalizing Concept Constructing questions Making observations Using available data Combining measurement operations 1. Constructing questions Close-ended (five choice) question – Respondents are offered explicit responses from which to choose. i. Advantages: o useful large sample, want to run statistics o Answer the question that researcher really wants them to answer o More response choices reduces ambiguity and makes it easier for respondent to answer, but only a certain amount is necessary o You know they will answer the question the way you want them to ii. Negatives: o Can obscure what people really think if the choices do not match the range of possible responses to the question o Can only be used to get at certain kinds of information Open ended questions – questions without explicit response choices; write in their own answer or verbal response i. Advantages o Useful when range of responses cannot adequately be anticipated o Useful to explore the meaning respondents give to abstract concepts ii. Negatives: o Sometimes the responses are difficult to compile and categorize… usually performed by a team 2. Making Observations Observations of characteristics of individuals, events, and places 3. Using Available Data (what we are doing) Readily accessible sources of social science data: i. Government reports ii. Law enforcement (union crime report) iii. Archives of social science surveys all hail the internet! Unlimited information available through a few simple clicks! Does the data set have the information you need to operationalize your variable? i. Avoid poorly constructed questions Have others used these indicators to measure this concept? How? i. Huge advantage ii. ICPSR: find and analyze data 4. Combining Measurement Operations Triangulation: the use of 2 or more different measures of the same variable i. Not the strangulation of a triangle, don’t worry. ii. Strengthens measurement: similar results with different measures of the same variable = evidence of validity Levels of measurement Nominal Ordinal Interval Ratio Why important? Better understand how cases vary on that variable and understand more fully what we have measured Has important implications for the type of statistics than can be used. Nominal Level of measurement Examples: nationality, occupation, or religious affiliation A, B, C, D Categorical or qualitative Can: distinguish difference Cannot: say one is “greater” than another, add, subtract, multiple, or divide Go ahead, try multiplying a nun and a monk and then subtracting them by a Baptist, let’s see what you get. They vary in quality but not in amount Special case of Dichotomies Dichotomies – variables having only 2 values i.e., gender (M or F) The level of measurement is nominal, but when a statistical procedure requires that variables be quantitative, a dichotomy can work Female (1), Not Female (0) Presence and absence of Greater and less than At the Quantitative Level Ordinal, interval, ratio Ordinal Level of measurement Examples: social class, level of conservatism, level of prejudice Underclass (1), working class (2), middle class (3), upper class (4) Not very important (1), fairly important (2), very important (3), most important (4) The numbers assigned to cases specify only the ORDER of the cases Can: distinguish difference, make greater and less than distinctions Cannot: add, subtract, multiply, or divide Index: a series of ordinal level measures may be combined into an index to assess a concept. For example: supportive parenting Interval Level of measurement Example: temperature (Fahrenheit) the numbers represent fixed measurement units but have no absolute, or fixed, zero point Can: distinguish difference, make greater and less than distinctions, add, and subtract Cannot: multiply or divide There are few true interval-level measures in the social sciences but we treat indexes as interval-level measurements Thursday, February 18, 2016 Ratio Level of measurement Example: people’s income The numbers indicating the values of a variable represent fixed measuring units and an absolute zero point (Zero point means absolutely no amount of whatever the variable indicates) Only level of measurement where the amount of space between each variable is fixed. Can: distinguish difference, be added and subtracted, make greater and less than distinctions, and because the numbers begin at an absolute zero point, they can be multiplied and divided. It is usually a good idea to try to measure variables at the highest level of measurement possible… The higher the level of measurement: The more info available, The more ways we have to compare cases And the more possibilities we have for statistical analyses Criteria for measurement quality Validity and Reliability Measurement Validity Extent to which an empirical measure adequately reflects the real meaning of the concept “… a valid measure of a concept is one that is closely related to other apparently valid measures of the concept, and to the known or supposed correlates of that concept, but that is not related to measures of unrelated concepts. Four Types (actually more) of Validity Face Validity Confidence gained from careful inspection of a concept to see if it is appropriate on its face Content Validity Does the measure cover the full range of the concept’s meaning? How is it achieved? Reviewing the literature thoroughly. Criterion Validity AKA Predictive Validity Comparing scores with those from a more direct or already validated measure Ex. A measure of blood-alcohol concentration could validate a self- report measurement of drinking Ex. A measure of attendance to religious services could validate a self-report measurement of religiosity. Construct Validity The degree to which a measure relates to other variables as expected theoretically Marital satisfaction and martial fidelity Measurement Reliability Is a matter of whether a particular technique, applied repeatedly to the same object, yields the same result each time Should produce consistent results Reliability is a prerequisite for measurement validity Four Possible Indicators of unreliability Test-Rest Reliability The degree to which two measurements of a phenomenon taken at two points in time are related to each other Interitem Reliability The questions should be highly associated with each other Cronbach’s alpha is a reliability measure commonly used to measure Interitem reliability Alternate Forms (split-half method) Reliability Reverses the order of the response choices in an index or modify the question wording in minor ways and readminister that index to subjects. If the two sets of responses are not different, then alternate- forms reliability is established. Interobserver Reliability Midterm 50 Multiple Choice Questions Chapters 1, 2, 4, 5, 6 from Babbie textbook 20 questions from lectures (I have lecture notes posted) 30 questions from book ( I have a study guide covering these) She likes lists and application questions, as well as definition questions.
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'