Types of Data

Question 1

How can you formulate a hypothesis?

Answer 1

• 1 problem
• 3 issues per problem (issues 1 - 3)
• 4 hypotheses per issue (hypothesis 1A - 1D)
• 4 questions per hypothesis (questions 1C-a - 1C-d)

Problem: people can’t support their waste

Issues: need to provide infrastructure; people not motivated; low awareness or instruction; laziness; lack of space;

Hypothesis - speculative statement

Question 2

What are issues?

Answer 2

Questions which need to be answered or topics which need to be explored in order to solve a problem

Question 3

What’s a hypothesis?

Answer 3

“A specified testable expectation about empirical reality that follows from a more general proposition; more generally, an expectation about the nature of things derived from a theory. It is a statement of something that ought to be observed in the real world if the theory is correct. See deduction and Chapter 3.”

Speculative answers for issues that are phrased as questions and/or exploration for issue phrased as topics

• Form hypotheses as a statement containing 3 basic components (two variables and an associative link between the two)
• EX: higher educated people have more awareness about recycling
• Not a problem if the relationship is opposite from what you thought or if there’s no relationship at all
• Serious problem if you’re not able to verify your hypotheses; you don’t have sufficient data to prove or disprove; then research essentially useless; can happen if you forget to add questions to your questionnaire

Question 4

What are Key Questions?

Answer 4

Questions that probe hypotheses and drive the research to solve the problem
Who will you count as people disposing of waste?

If one concept is complex, you could have a whole list of questions

Question 5

What are concepts?

Answer 5

Concept is abstraction/representation of an object or a behavioral phenomenon

• There’s something like social reality and you can’t affect natural social processes; attitudes are the social fact; what you might do is find a way to indicate that, or how
• You may have two different researchers with two different sets of results on the same problem
• You must develop and identify your goals; must have well-targeted questions; this distinguishes good research from bad

Question 6

Why do we need concepts?

Answer 6

Concepts provide a common language which enables researchers to communicate

• Each discipline has differernt types of concepts
• • Psychology - depression, conception, learning
• • Poli Sci - power, democracy, regime
• Essential to be aware of the important concepts for your field

Conceptual definitions are the definitions that describe concepts by using other concepts

Example:
Rich people vote for right-wing parties (variables: rich people & political parties)

You may ask about:
– Income, education, investments, career, weekly spending, lifestyle, preferred stores/brands, assets,

• 15-25% of people won’t respond to questions about income (average non-response 1-2%)
• Education often an indicator of income, but that relationship is not linear (teachers & researchers vs. doctors & lawyers)
• May have low income but high assets
• Weekly spending or investments may reveal available resources
• Heirs to fortunes - no education, income, etc, but huge assets

Results of your study will be different depending on whether you’re asking the right questions

Important to think of concept in details

Question 7

What’s conceptualization?

Answer 7

“(1) The mental process whereby fuzzy and imprecise notions (concepts) are made more specific and precise. So you want to study prejudice. What do you mean by “prejudice”? Are there different kinds of prejudice? What are they? See Chapter 6, which is all about conceptualization and its pal, operationalization.
(2) Sexual reproduction among intellectuals.”

The process through which we specify what we mean when we use particular terms in research

Each respondent must understand terms in the same way; need to be specific

• Might even ask: are you rich?
• You must ask questions that are narrowing the fuzziness and narrow scope of how people might define “being rich”
• Must ask questions that are not so sensitive yet still reveal sensitive information (too sensitive might produce lies or no response)

Question 8

What’s an object?

Answer 8

Refers to a tangible item in a person’s environment that can be clearly and easily defined

Each object has
- Objective properties - directly observable; (ex: age, number of purchases, marital status)

• Subjective properties - intangible, abstract; (ex: attitudes, feelings, expectations, perception)

Question 9

What’s a variable?

Answer 9

“Logical sets of attributes. The variable sex is made of up of the attributes male and female. See Chapter 1.”

Variable is any entity that can take on different values; anything that can vary is considered as variable

Each variable must be exhaustive, it should include all possible answerable responses/attributes

• Variable “Religion”
• • 1 = Protestant
• • 2 = Jewish
• • 3 = Muslim

Open vs closed questions, order in questionnaire, the phrasing of question - all might increase/decrese chances of reliable and unbiased answers

Question 10

What are attributes?

Answer 10

“Characteristics of people or things. See variables and Chapter 1.”

For variable “Colors”, attributes are blue, purple, green, etc.

Must be mutually exclusive, ex: no respondent must not have two responses simultaneously (no overlaps)

Variable “age”
1 = less than 18
2 = 18-30
3 = 30-40
4 = 40 - 50
5 = more than 50
-- Mistake because numbers overlapping

Variable “agreement”
1 = strongly agree
2 = agree
3 = disagree
4 = strongly disagree
-- Formally ok because there’s no overlap; but no midpoint (matter of debate); even/odd number of questions also debatable

Question 11

What are Definitions? (2 types)

Answer 11

1. Normal definition
   = is simply assigned to a term without any claim that the definition represents a “real” entity
   Not used in research or when defining variables
2. Operational definition
   = specifies precisely how a concept will be measured - that is, the operations we will perform
   – In research, extremely helpful to clarify things
   – Ex: Height - measurement from the heel of your foot to the top of your head

Question 12

Concept of Measurement

Answer 12

Measurement
= standardized process of assigning number or other symbols to certain characteristics of the objects of interest

= careful and deliberate observation of the real world for the purpose of describing objects in terms of attributes composing the variable

Researchers engage in assigning numbers or labels to:

• • People’s thoughts, feelings, behaviors, and characteristics
• • Features or attributes of objects
• • Aspects of concepts/ideas

Generally applied to all aspects of research

Question 13

4 types of measurement

Answer 13

Levels of measurement refer to the relationship among values that are assigned to the attributes for a variable

There are typically 4 levels of measurement

1. Nominal
2. Ordinal
3. Interval
4. Ratio

Guideline of how to measure properly and abstract info from the real world

Question 14

Nominal Measurement

Answer 14

“A nominal variable has attributes that are merely different, as distinguished from ordinal, interval, or ratio measures. Sex is an example of a nominal measure. All a nominal variable can tell us about two people is if they are the same or different. See Chapter 6.”

At the nominal level of measurement, numbers are assigned to a set of categories for the purpose of naming, labeling, or classifying the observations

Basic example of classification

Example “gender”
1 = male
2 = female

No logic behind numbers behind differentiation as long classification is consistent

Creates more work if you don’t display numbers but if they would distract attention of respondents, still perhaps better to hide the numbers

Question 15

Ordinal Measurement

Answer 15

“A level of measurement describing a variable with attributes we can rank‐order along some dimension. An example is socioeconomic status as composed of the attributes high, medium, low. See also Chapter 6 and interval measure, nominal measure, and ratio measure.”

In ordinal level of measurement the attributes can be rank-ordered. Distances between attributes do not have any meaning

Example “satisfaction”
1 = very satisfied
2 = somewhat satisfied
3 = somewhat dissatisfied
4 = very dissatisfied

Transitivity must be met: a > b > c > d
Ordinal = order

As long as numbers are following order, you could change scale to:
4 = very satisfied
3 = somewhat satisfied
2 = somewhat dissatisfied
1 = very dissatisfied
– More logical order because you’ll have a more accurate mean score

Distance between variables could be any measure; we don’t know

Question 16

Interval Measurement

Answer 16

“A level of measurement describing a variable whose attributes are rank‐ordered and have equal distances between adjacent at‐ tributes. The Fahrenheit temperature scale is an example of this, because the distance between
17 and 18 is the same as that between 89 and 90. See also Chapter 6 and nominal measure, ordinal measure, and ratio measure.”

In interval measurement, the distance between attributes does have meaning
– No absolute zero; must gauge based on ratio

Example “Temperature (in Fahrenheit)”
– Can’t say “80 is twice as hot as 40”

Example “IQ”

Question 17

Ratio Measurement

Answer 17

“A level of measurement describing a variable with attributes that have all the qualities of nominal, ordinal, and interval measures and in addition are based on a “true zero” point. Age is an example of a ratio measure. See also Chapter 6 and nominal measure, interval measure, and ordinal measure.”

In ratio level of measurement, there’s always an absolute zero that is meaningful, i.e. you can construct a meaningful fraction (or ratio)

• • Example “age”
• • Example “children”

Question 18

Hierarchy of Measurements

Answer 18

It’s important to recognize that there’s a hierarchy implied in the level of measurement idea

Order from Best - Worst:

1. Ratio -> Absolute Zero
2. Interval -> Distance is meaningful
3. Ordinal -> Attributes can be ordered
4. Nominal -> Attributes are only named

• If you can phrase a question in a higher ordinal level, you should
• Always start with ratio valuables

You can always simplify into lesser categories later, but you can’t go upwards afterwards
– Ex: if you ask for age (i.e. 24) you can put into categories later, but if you ask for age categories (i.e. 20-30) those results will always be vague

• Don’t ask level of support, but ask willingness to pay (they’ll express in currency rather than words)
• Results in highest level of data rather than vague

At lower levels of measurement, assumptions tend to be less restrictive and data analyses tend to be less sensitive

At each level up the hierarchy, the current level includes all of the qualities of the one below it and adds something new

In general, it’s desirable to have a higher level of measurement (e.g. interval or ratio) rather than a lower one (nominal or ordinal)

If you know at the end that you need to prove your hypotheses, you know that variable must be at ratio level and design variable in a way to get that type of measurement; if you take only ordinal, you won’t be able to verify your hypothesis

Question 19

Why is level of measurement important?

Answer 19

First, knowing the level of measurement helps you decide how to interpret the data from that variable

Second, knowing the level of measurement helps you decide what statistical analysis is appropriate for the values that were assigned

Statistical tools won’t guide you here

• • You can’t take a mean from nominal variables (gender = 1.14?)
• • For things like satisfaction, you might be able to take a mean but then you must be very careful about how you interpret the data