SOC221H5 Chapter Notes  Chapter 14: Box Plot, Categorical Variable, Squared Deviations From The Mean
by OC590131
Department
SociologyCourse Code
SOC221H5Professor
David P.Chapter
14This preview shows pages 13. to view the full 10 pages of the document.
SOCIAL STATISTICS
For a Diverse Society 8th Edition
Soc222 Textbook Notes
CHAPTER 1
Stats: set of procedures used by social scientists to organize, summarize, and communicate
numerical information
Data: info represented by numbers
The Research Process
1. Asking research questions
 Empirical research based on information that can be verified by using our direct
experience

Theory: a set of assumptions and propositions used by social scientists to explain, predict, and
understand the phenomena they study.
 Attempts to establish a link b/w data and the understanding as to why things are
related to each other in a certain way
2. Formulating the hypothesis
 Hypothesis: predicting the relationship b/w two or more observable attributes
 Are verified after they have been tested empirically
 Not all are derived directly from theories, can be from observations or from intuition
 A statement of a relationship between two characteristics (variables)
 Hypotheses are statements written in a special language that describe the relationship
between two variables.
 You can obtain two pieces of information from a wellwritten hypothesis:
 1. the dependent and independent variables.
 2. the direction of the relationship between these two variables.
Variable: property of ppl or objects that takes on two or more values
 I.E. people can be classified into social class categories upper, middle, and lower
class
 Social scientists must choose a unit of analysis, which is the object of their research
Independent and dependent variables: Causality
Dependent: the effect
Independent: the cause
Variables are influenced by other variables.
– That is, the attributes that one variable takes, the DEPENDENT (y) variable, are
influenced by the attributes of another variable, the INDEPENDENT (x)
variable.
 To establish that two variables are causally related the analysis must meet conditions
find more resources at oneclass.com
find more resources at oneclass.com
Only pages 13 are available for preview. Some parts have been intentionally blurred.
SOCIAL STATISTICS
For a Diverse Society 8th Edition
Soc222 Textbook Notes
1. The cause has to precede the effect in time
2. There has to be an empirical relationship between the cause and effect
3. The relationship can’t be explained by other factors
Guidelines:
1. The dependent variable is always the property that you are trying to explain; its always
the object of the research
2. The independent variable usually occurs earlier in time than the dependent variable
3. The independent variable is often seen as influencing, directly or indirectly, the
dependent variable
The purpose of the research will help determine which is the indp. And dep. Variable
3. Collecting Data
 Measuring variables and collecting data
 i.e. questionnaires
NOMINAL LEVEL OF MEASUREMENT
 numbers or symbols are assigned a set of categories for the purpose of naming, labelling, or
classifying observations. I.e. Gender, separating in categories where 1 represents males and 2
 nominal variables are called qualitative because different categories vary in the
quality inherent in each, not quantity.
 Nominal variables should include categories that are both exhaustive and mutually
exclusive
Jointly Exhaustive
 there should be enough categories composing the variables to classify each observation i.e. by
adding category “other”.
• Variable attributes should have certain features.
• Jointly Exhaustive means that every observation in your sample should have a place and
they can be classed within the variable.
Let’s say you have the categories: professional, manager, skilled worker, unskilled worker.
This will create data problems since your systematically (bias)
excluding people which have a certain attribute.
(i.e. the woman playing golf, maybe she’s unemployed or retired!
Mutually exclusive
 one category suitable for each observation
• Mutually exclusive means there is only one category suitable for each observation.
– In other words, you can’t have overlapping attributes or categories.
I.E
find more resources at oneclass.com
find more resources at oneclass.com
Only pages 13 are available for preview. Some parts have been intentionally blurred.
SOCIAL STATISTICS
For a Diverse Society 8th Edition
Soc222 Textbook Notes
Annual Household Income
Less than $9,999
$10,000 to $19,999
$20,000 to $39,999
$40,000 to $59,999
$60,000 to $79,999
$80,000 to $99,999
Greater than $100,000
Nominal variables that have only two attributes (often gender is presented that way) are called
dichotomous. You fit into one category or the other that is that. i.e. employed or unemployed,
married or not married
 have the least amount of information and you cannot convert these to intervalratio.
ORDINAL LEVEL OF MEASUREMENT
when you assign numbers to a rank ordered category ranging from low to high
should also include categories that are mutually exhaustive and exclusive
Also qualitative
Most attitude measures are ordinallevel measures.
i.e. “agree”, “somewhat agree”, etc.
INTERVAL RATIO LEVEL OF MEASUREMENT
if the categories of a variable an be rank ordered and if the measurements are expressed in the
same units, and equally spaced
can compare values in terms of which is larger but also how much larger one is compared to
another
variables with a natural zero point are called ratio variables
• the value of 0 is not arbitrary – it actually means 0. i.e. temperature
interval ratio level you have the most info. So you can always convert a variable that is interval
ratio to nominal or ordinal
This is a quantitative level of measurement, and unlike ordinal level, the distance between two
NUMBERS is KNOWN.
Discrete Variables
 have minimum sized unit of measurement, which can’t be subdivided i.e. # kids per
fam, where min unit is 1 kid. A fam might have 2 or 3 kids, but never 2.5
find more resources at oneclass.com
find more resources at oneclass.com
You're Reading a Preview
Unlock to view full version