SOCI 2030 –Chapter 6 – 10
Chapter 6 - Qualitative and Quantitative Measurement:
− What is it?
− The process of creating measurable concrete variables from abstract concepts.
− E.g How Parental divorce influences children’s wellbeing?
− You need to think of something tangible – you have to make sure you find ways to
− When you define “wellbeing” do you mean mental health, behavior at school, resources
at home, physical, social and intellectual wellbeing. Wellbeing on its own is very
abstract. When you find out what you are going to focus on, you have to find ways to
− Ask questions about (e.g. Mental Health – ask Psychological questions)
− Defining your concept first then measure concept.
− Concept: Ideas or Mental representations of things. (eg. Crime, Gender, Alienation,
Love, Life Satisfaction)
− Concepts may be independent or dependent variables.
− We use concepts to develop hypothesis etc. These are not easily observed.
− In social research, we define concepts in 2 different ways:
1. Nominal: Describes the concept in words. (e.g. “Political Part Identification”: the party to
which people most closely associate themselves.
− The dictionary definition
2. Operational: Describes how the concept is to be measured.
− E.g. “Political Party Identification” may be measured by asking people, “Do you
normally think of yourself as a supporter of the Conservatives, Liberals, New
Democrats, Greens or Block Quebeccois?”
− Could ask which party did they voted in the last election.
− Think about how you are going to measure.
SOCI 2030 – Ch 6 – 10 Review Page 1 The Measurement Process:
− Quantitative = Deductive
− Qualitative = Inductive
− Both involve conceptualization and operationalization
− More of a concern for quantitative researches.
Steps - Quantitative:
1. Take a General Concept or Idea
2. Specify the dimensions that you want to study; develop conceptual definition
(Conceptualization – Defining Concept – Define what you want to do, what you want to
3. Create measures to evaluate these dimensions (Operationalization – you find the
measures, what kind of measurements, how are you going to conduct your research)
Process of Qualitative Operationalization:
1. Empirical Observations (Operationalization)
2. Working Ideas (Conceptualization)
4. Generalizations / Theories
− Consistency or dependency of a measure – Does the measure consistently give the
same results? – If not, there might be an issue.
− Clear conceptualization
− Use of most precise (or highest) level of measurement possible
− Use multiple indicators (Don’t just ask one question, ask multiple questions in different
ways – make people answer it twice)
− Use of pretests, pilot studies, replication (Research with a smaller group of people,
replicate your study. Pilot study – before doing real study, you do a smaller “test driver” if
your questions work, how long it takes, it there are any problems etc)
− Truthfulness of a measure – is it measuring what the researcher thinks it is measuring?
− Was the correct test used?
SOCI 2030 – Ch 6 – 10 Review Page 2 − Can alternative explanations for change in the dependent variable be eliminated? (You
should try to eliminate alternative understandings)
Criteria of Measurement Quality: Validity and Reliability:
− Reliable but not Valid (Picture of a dart board, the answers are relevant but it’s not in the
− Valid but not Reliable (Some are in the bulls eye, and some are spread out. They are
− Valid and Reliable (You get the results all similar and getting reliable results)
Variables and Attributes:
− Variable: Gender, Age
− Attribute: Female, Old
− Variable: Agreement
− Attribute: Strongly Agree, Agree, Neutral, Disagree, Strongly Disagree etc.
Variables should have 2 qualities:
1. Attributes must be exhaustive (We are talking about measurements, how we measure a
variable – how we ask questions. When you ask a question, you should include all
2. Attributes must be mutually exclusive. (Cannot have more than 2 answers. E.g. Age. You
cannot be old and young. They should also put all possible attributes here.)
Level of Measurement:
− Attributes of the variable are merely different; Exhaustive and Mutually Exclusive
− Ex. Gender (Male and Female)
− Religious Affiliation, College Major, Hair Colour, Birthplace, Nationality
− You cannot say one attributes work better than the other. They are just
characteristics, there is no order. Nothing is better or worse.
− Variables with attributes we can logical rank in order.
− E.g. Socioeconomic status (High/Middle/Low), level of Conflict, Conservativeness
SOCI 2030 – Ch 6 – 10 Review Page 3 − Variables whose attributes are rank – ordered and have equal distances between
− E.g. Temperature (Fahrenheit) … there is an order, hotter or colder. There is
equal distances, IQ score (Levels of IQ, there is a difference – equal distances)
− There is another intervals
− Doesn’t have a true zero value – they are all numbers.
− Variables whose attributes meet the requirements of an interval measure and
HAVE A TRUE ZERO point.
− Example: Age (0,1,2 ….), Length of time, Number of Organizations, Number of
Groups, Number of as Received in College
− Ratio is more common
Chapter 7 – Qualitative and Quantitative Sampling:
− Selection of a subset of individuals from a population
1. Non – Random Sampling:
− Not representative of Population
2. Random Sampling:
− Representative of Population
− We need to plan how we are going to choose our sampling method in our research
What is the Logic of Sampling?
− We want to talk about the population (Woman, Refugees etc.)
− Think who you are going to sample, think what is possible, what is accessible, create a
sampling frame – sometimes you are not able to get everyone in your SAMPLING
SOCI 2030 – Ch 6 – 10 Review Page 4 FRAME. Sometimes you end up with people less than your sampling frame. (Sampling
Frame – people you think can gain information and data from. Who you actually talk and
survey is your sampling group.)
− If you are conducting quantitative research, you want to make generalizations of the
− If you are not a quantitative researcher, you don’t think about generalization, you are not
concerned about looking for a representable symbol. You aren’t aiming to talk about the
Non Probability Sampling:
1. Convenience (Haphazard Sampling):
− Reliance on available subjects
− Also called accidental sampling
− Rely on available subjects, you just go and gather information from people who
are convenient. You start with people you know – family, friends etc.
2. Purposive or Judgmental Sampling:
− Use expert judgment to pick cases
− Making judgments to who you will pick and who you are going to talk with.
− Sometimes you think that you are going to study this, but instead of sampling
people – you ask the expert of the topic
− E.g. Diversity in Workforce (Asking TD (A very Multicultural Bank, etc)
3. Snowball Sampling:
− Network or Chain Referral
− You question someone and you ask if they have another person that they are
willing to introduce to you to another sampler. You start with a few people and
ask for sampling.
− E.g. Cancer Patients, if you want to study cancer conditions, you can ask for help
to reach these people.
4. Quota Sampling:
− Establish categories of cases
SOCI 2030 – Ch 6 – 10 Review Page 5 − Choose fixed number in each category
− You want to include everything, all characteristics of a category
− You want to get all groups – you don’t want to exclude any groups.
− Used for Quantitative Research
− Representative Population
− Representativeness – the quality of a sample of having the same distribution of
characteristics as the population from which it was selected.
− Can generalize from sample to population
− You want to talk about the sample at the end.
− Choose a sample and use information gather from the cases in the sample to generalize
to the population.
− You want to have a variety of population – you do not want the population so big
because it won’t be homogenous.
How can we ensure samples are representative?
− The Tule of Epsem (Equal Probability of Selection Method): All members of the
population have an equal chance of being selected in the sample.
− Statistics: Are mathematical characteristics of samples
− Parameters: Are mathematical characteristics of populations
− Statistics are used to estimate Parameters
1. Simple Random Sampling (SRS):
− 1. The Units composing a population are assigned numbers
− 2. A set of random numbers is generated – Included people that assigned these
− 3. The units having those numbers are included in the sample.
SOCI 2030 – Ch 6 – 10 Review Page 6 2. Systematic Random Sampling:
− The Sampling starts by selecting an element from the list at random and then
every “kth” element in the frame is selected.
The Ethics of Sampling:
− Probability Sampling: A risk of error. Inform readers of any errors that might
make results misleading.
− Non Probability Sampling: Ensure that readers do not confuse variation with
that’s typical in the population.
Chapter 8: Survey Research:
− A very old and popular method to collect information about a population
− Useful when:
1. Collecting Original Data (First time you are asking these questions to these
people = you are not using something that is available. It is collected by you)
2. Describing a large population
3. Measuring attitudes and orientations
− Unit of Analysis – respondents (Individual who answers your questions, who fills up your
− Systematic questionnaire or interview. Involve filling out a form which is then returned to
− Documents containing questions and other types of items designed to solicit information
− You can see them as structured interviews without an interviewer.
− Answers are usually really structured – you need to pick from the answers that are given
Types of Survey Research:
1. Mail Surveys:
− Questionnaires sent by mail
− Mail distribution and return practices (Put in prepaid envelopes)
− Rewards for filling out the survey
SOCI 2030 – Ch 6 – 10 Review Page 7 − Monitoring Returns – see how many are sending them back to you. See the
− Follow up mailings (Check to see if they replied, if they didn’t – will try to send
− Response rates: 50% adequate, 70% very good.
− Easy to discard – be short
− Had to be very clear and easy to follow
2. Face to Face Interview Surveys:
− Interview Training – if you get someone to interview for you; you have to train
your employees to interview people. Training – explain what you are trying to
find, explain all the questions and what they have to do. Make sure they are all
on the same page)
− Familiarity with questionnaire – know what you are going to ask.
− Interview schedule – schedule interview, keep recording for future use and the
planning. Use Pseudo names.
− Most follow question wording exactly. You have to be consistent. You want to
make sure all your interviewers are following the question and collecting the data
− Must record responses exactly.
− Also record other events during Interview
− Appearance and demeanor crucial – we all judge people on their appearances
and how we behave.
3. Telephone Surveys:
− Questions are asked over the phone
− Can reach 95% of households
− Random Digit Dialing (RDD) – Calling every number or randomly dialing
− Selecting respondents from a household.
− Computer – Assisted telephone Interviewing (CATI)
− Interviewers sit at a Computer Terminal
− Answers directly entered into the computer
SOCI 2030 – Ch 6 – 10 Review Page 8 − Very Fast
− Often used in Polling
4. Internet (Web) Surveys:
− Used of Email or Web
− In US – 69.6% of population now use internet
− Canada – 67.5% of population (2007)
− Useful features such as filter questions, drop-down menus ete.
1. Question and Hypotheses (Make sure you have at least one hypothesis per variable
2. Existing scales or measures (Can Modify)
3. Avoid double-barreled questions (Do you favor legalization of marijuana for use in
private homes but not in public spaces?)
− You should ask these separately, it would be confusing for you and the
4. Avoid Assumptions (e.g. How many children do you have?)
5. Avoid biased items and terms (Doctors believe etc etc etc. Then the respondent would
probably believe what the doctor said)
6. Choose Clear and Sensitive Wording – You should make sure everyone can
understand what you are saying; different literacy levels. And be sensitive, some people
are extremely sensitive to some topics.
7. Avoid Ambiguity and Vagueness: Do not use the word “regularly”, “often”; be very
clear with your questions, some people may interpret it differently.
8. Avoid Jargon, abbreviations and emotional language – E.g. Talking about probability
sampling: you do not know that the person would understand.
9. Avoid Negatives and Double Negatives – (Are you not unconvinced by this
argument?) – These statements and questions are confusing.
10. Don’t use leading questions (Don’t ask “You don’t smoke, do you?”
11. Try to not ask questions beyond a respondent’s capabilities
12. Use of Mutually Exclusive Categories
13. Use of Exhaustive Categories – Provide all possible options.
SOCI 2030 – Ch 6 – 10 Review Page 9 − Scales/ Indexes on attitudes, believes of behaviors. (Putting a chart or research and ask
the questions. And cite- it resources)
− Open-Ended Questions:
− Where do you currently live?
− How many children do you have?
− Try to give, close ended questions – so that you can analyze your data easier and you
have the categories – you list the attributes and you tell the person to pick the
category. Provide categories and let them pic