CHYS 2000 - Quiz Flashcards
Quantitative Measurement and Concepts in Social Research
Quantitative measurement is a distinct step in the research process that requires special terminology.
Concepts (referred to as variables in quantitative measurement) are established as numerical data.
Recall – Quantitative measurement is a deductive process:
(1) Establish the abstract concept.
(2) Identify the measurement procedure.
(3) Research the empirical evidence to (dis)prove the consistency of the concept.
Concepts in Social Research
Concepts – An idea expressed as a symbol or in words.
Unemployment is a concept, unemployment rates is a way to measure the concept.
Quantitative Measurement and Concepts in Social Research
Conceptualization – The process of developing clear, rigorous, systematic conceptual definitions for a concept. -thinking about it
Conceptual definition – A careful, systematic definition of a construct that is explicitly written to clarify one’s thinking – This is exactly what I mean by this concept in this context.
Conceptualization
The process of developing clear, rigorous, systematic conceptual definitions for a concept. -thinking about it
Conceptual definition
A careful, systematic definition of a construct that is explicitly written to clarify one’s thinking – This is exactly what I mean by this concept in this context.
Operationalization
The process of moving from the conceptual definition of a concept to a measurement of that concept which allows the research to observe it empirically.
Operational Definition
The precise measurement of a variable through the use of empirical evidence.
Quantitative Measurement and Concepts in Social Research
Example
Concept – Childhood
Operationalization of Childhood: Do you agree with the following conceptualization of childhood; By childhood, I am referring to all individuals from birth to eighteen years of age who attend school regularly and are looked after by a parent/guardian.
Likert Scale:
Strongly Agree / Agree / Disagree / Strongly Disagree
1) Nominal Measures
– The lowest, least precise level of measurement (there are only differences in types among the categories of the variable. (not ranking)
(i.e.) Religion – Islam, Judaism, Christianity, Buddhism
Ordinal Measures
Identifies a difference among categories of a variable through rank order.
(i.e.) Likert Scale - Strongly Agree, Agreed, Disagree, Strongly Disagree
3) interval Measure
Identifies differences among variable attributes but there is no true zero.
(i.e.) Celsius Temperature - - 5 degrees, 10 degrees, 15 degrees
(4) Ratio Measures
The most precise level of measurement where a variable attribute can be ank ordered, the distance between each is precise, and absolute zero exists.
(i.e) money – $0, $10, $100
Quantitative Measuremen
Hypothetical Causal Relationship
The researcher expresses the relationship among variables in a theoretical manner (prior to research).
Conceptualization (including the definition)
Operationalization (including the definition)
Tested Empirical Hypothesis
The researcher expresses the relationship among variables in an empirical way (after conducting research).
Quantitative Measurement and Concepts in Social Research – Conceptualization and Operationalization
1) Abstract Concept
Social Networks
(2) Conceptual Definition
By the refugees social networks, I am referring to the people they are connected with.
(3) Operational Definition
An index to determine how many members are a part of the refugees social network (i.e. 0 – 20 family members, i.e. 0 – 20 work members, 0 – 20 community members).
(1) Abstract Concept
Job Quality
(2) Conceptual Definition
By job quality, I am referring to the desirability of the refugees’ job.
(3) Operational Definition
An index of job quality includes (a) the status of the job (i.e. high or low), (b) if the job was permanent or temporary, and if the refugees’ educational qualifications matched
Hypothetical Causal Relationship
The social networks of recent refugees affect their quality of employment in Canada. -mpc/
Tested Empirical Hypothesis
The greater the social networks of recent immigrants, the more likely they are to have higher-quality jobs.
-mpc/
Quantitative Sampling
Sample – A smaller set of cases a researcher selects from a larger pool and generalizes(nomothetic, objective ) to the population.
Probability Sampling – Sampling strategies that give every subject in the population the same likelihood of being selected – this means it is a random process.
Sampling is about extrapolation – For instance, how do you make a sample of 2000 people account for the general positions of 2 million people?
Quantitative Sampling – Probability Sampling in Social Research
Population – The large group of cases from which a researcher draws their sample (i.e. university students).
Sampling Frame – A list of cases in the population (i.e. university registrar list of students).
Sample – A smaller set of cases a researcher selects from a larger pool and generalizes to the population.
Sampling Ratio – The number of cases in the sample divided by the number of cases in the sampling frame.
If the sample frame (university registrar) has 5,000 students, and a researcher draws 150 students (sample) from it, the sampling ratio is 150/5,000 = 0.03 or 3 percent.
(1) Simple Random Sample
Researchers create a sampling frame and use a pure random process to select cases.
Population – 5,000 university students
Parameters – Determine the average age of the students (age varies between 17-55).
Randomly select 50 students – 50/5,000 = 0.1 percent sample
(2) Systematic Sampling
Researchers select every x case in the sampling frame using a sampling interval.
Sampling Interval – The sampling interval (i.e. 1 in x, where x is a number) tells the researcher how to select elements from a sampling frame.
If one wants to sample 300 names from a list of 900, choose a random starting point, select every third name (the sampling interval is 3), and you will generate 300 samples from the 900 in the sampling frame.
(3) Stratified Sampling
Researchers identify a set of mutually exclusive and exhaustive categories, then use a random selection method to select cases for each category.
Steps in Stratified Sampling
(1) divide the population (10,000 employees) into sub-populations or strata.
Able-bodied persons (92%) anbd persons with exceptionalities (8%).
(2) Create a sampling frame for each strata.
9,200 employees and 800 employees
(3) Decide on a sample size and randomly select from each strata.
How did a question related to work-place support for persons with exceptionalities differ among able-bodied persons and persons with exceptionalities?
(4) Cluster Sampling
Researchers use multiple stages and cover a wide geographic area where samples are randomly selected from.
Cluster – A unit that contains final sampling elements but can be treated temporariliy as a sample element itself.
4-Stage Cluster Sample
Step 1 – Mapleville has 55 districts. Randomly select 6.
Step 2 – Divide each district into blocks. Each district contains 20 blocks. Select 4 blocks from each district.
Step 3 – Divide blocks into households. Each block as 45 households. Select 10 households from each block.
Step 4 – Select a respondent from each household.
(5) Random Digit Dialing
Researchers select cases for telephone interviews that use all possible telephone numbers as a sampling frame.
The population is telephone numbers.
Canadian telephone numbers (area code) (active exchange number) (four-digit number in exchange).
905 555 0000
905 555 0001
905 555 0002
905 555 0003
Survey Research
A quantitative social research technique in which one systematically asks many people the same questions, then records and analyzes the answers.
Survey Research
Surveys ask people (known as respondents) about:
Behaviour – When did you last visit a close relative?
Attitudes/Beliefs/Opinions – What is the biggest problem facing Canada today?
Characteristics – What is your age?
Expectations – Do you plan on attending university?
Self-Classification – Do you consider yourself liberal, moderate, or conservative?
Knowledge – Who was elected prime minister in the last election?
Researchers warn against asking ‘why’ questions because it requires more elaboration than a survey can provide.
mpc question
Writers should avoid
Jargon, slang, abbreviations, ambiguity, confusion and vagueness
Prestige bias: A problem that occurs when a highly respected group/individual is linked to one of the answers.(the influence of someone else)
Double-barreled questions: A problem when two ideas are combined into one question (does your employer have pension and health insurance benefits? – what if it’s one and not the other)
Leading questions: A question that leads the respondent to choose one response over another (you don’t smoke, do you?’ – one may say they don’t even though they may.
Asking questions beyond a respondent’s capabilities.
Threatening Question
Respondents are less likely to convey their true behaviour or beliefs because they fear a loss of self-image.(ex house hold income)
Questions pertaining to income do not always generate a response because people find them intrusive.
Socially Desirable Questions
A bias in survey research in which respondents give a socially acceptable answer rather than giving a truthful answer.
People tend to over-report being cultured, giving money to charity, and volunteering.
Knowledge Questions
Researchers sometimes want to find out whether respondents know about an issue but knowledge questions can be threatening because respondents do not want to appear ignorant.
Studies suggest a large majority of the public cannot correctly answer elementary geography questions.
Open Ended Questions
Respondents are free to offer any answer they wish to the question.
(i.e.) What is your favourite television show?
Closed Ended Questions
Respondents must choose from a fixed set of questions.
(i.e.) Is the Prime Minister doing a good job?
Strongly agree
Agree
Disagree
Strongly disagree
Skip or Contingency Questions
A two or more part question where the answer to the first determines which of two different questions a respondent receives next.
A contingency question on drinking alcohol can immediately redirect respondents who never drink as the following questions are irrelevant to them.if yes to the first question then jump to question 5
Experimental research is informed by the positivist paradigm
At its core, an experiment is when you modify something in a situation and then compare the outcome to what existed prior (without the modification).- is causation
Experimental Research is a three-fold process:
(1) Begin with a hypothesis
A person tries to start their car. It does not start. One experiments by cleaning the battery and trying again.
(2) Modify the situation
The person modified something (cleaned the connection).
(3) Compare the outcome with the modification
Compare the outcome (whether the car started) to the previous situation (it did not start).
A buildup of crud on the connection is why the car is not starting,. Once the crud is off, the car will start.
Experimental Research’s strength lies in its ability to demonstrate causality.
(1) temporal Order
(2) Association
(3) No alternative explanation
In experimental research, one can actually observe causation. In other forms of research, causation is implied.
mpc
The Classical Experimental Research Process
(1) Random Assignment – Dividing the collection of cases into two or more groups through a random process.
(2) Experimental Group – The group that receives the treatment in experimental research.
(3) Control Group – The group that does not receive the treatment in experimental research.
(4) Pre-Test – The measurement of the dependent variable of an experiment prior to the treatment.- to determine how people are timpacted by the dv
(5) Treatment / Independent Variable / No Treatment – Researchers measure the impact of the independent variable by creating a condition or situation while the control group does not receive the independent variable (they receive a placebo).
(6) Outcomes / Dependent Variables – The physical conditions, social behaviours, attitudes, feelings, or beliefs of subjects that change in response to a treatment.
(7) Post-Test – The measurement of the dependent variable in experimental research after the treatment.