Chapter 1 Flashcards
why learn statistics?
First, you are
constantly exposed to statistics every day of your life. Marketing surveys, voting polls, and
social research findings appear daily in the news media. By learning statistics, you will
become a sharper consumer of statistical material. Second, as a major in the social sciences,
you may be expected to read and interpret statistical information related to your occupation
or work.
what is stats
But
the word statistics also refers to a set of procedures used by social scientists to organize,
summarize, and communicate numerical information.
what kind of info can b the subject of stat analysis
info represented by numnbers ONLY
how to think of research process?
. We can think of the research process as a set of activities in which
social scientists engage so that they can answer questions, examine ideas, or test theories.
research process studies
. Asking the research question
- Formulating the hypotheses
- Collecting data
- Analyzing data
- Evaluating the hypotheses
. For example, the questions “Is racial equality good for society?” and “Is an
urban lifestyle better than a rural lifestyle?” cannot be answered empirically - why
because the
terms good and better are concerned with values, beliefs, or subjective preference and,
therefore, cannot be independently verified
how to come up w a research q?
pick a q that interests u. u can find inspo all around u
what relationship lies at the heart of social sci inquiry
The relationship between attributes or characteristics of individuals and groups
lies at the heart of social scientific inquiry.
diff btwn theories in life n in sci?
y. Whereas our theories about events
in our lives are commonsense explanations based on educated guesses and personal
experience, to the social scientist, a theory is a more precise explanation that is frequently
tested by conducting research.
what is a theory and its goal?
A theory is a set of assumptions and propositions used by social scientists to explain,
predict, and understand the phenomena they study.
4 The theory attempts to establish a
link between what we observe (the data) and our conceptual understanding of why certain
phenomena are related to each other in a particular way.
after coming up w a theory n qs, what is the next step
Our next step is to test some of the ideas suggested by the gender segregation theory
what do theories suggest
. Instead, theories suggest specific concrete
predictions or hypotheses about the way that observable attributes of people or groups are
interrelated in real life
why are hyporthesis tentative and what are they
Hypotheses are tentative because they can be verified only after they
have been tested empirically. A statement predicting the relationship between two or more observable attributes.
are all hypotheses derived from theories and if not, where can they come from
Not all hypotheses are derived directly from theories. We can generate hypotheses in many
ways—from theories, directly from observations, or from intuition. Probably, the greatest
source of hypotheses is the professional or scholarly literature
what is a unit of analysis
Social scientists must also select a unit of analysis; that is, they must select the object of
their research. We often focus on individual characteristics or behavior, but we could also
examine groups of people such as families, formal organizations like elementary schools or
corporations, or social artifacts such as children’s books or advertisements
If we examine how often women are featured in prescription drug advertisements, what is the unit of analysis
the ads
how are hypotheses usually stated
d in terms of a relationship between an independent and a
dependent variable.
what is the variable the researcher wants to explain
dependent
what is an inependent vble
The variable that is expected to “cause” or account for the
dependent variable is called the independent variable.
how can you say two variables are causally related
To establish that two variables are causally related, your analysis must meet three
conditions: (1) The cause has to precede the effect in time, (2) there has to be an empirical
relationship between the cause and the effect, and (3) this relationship cannot be explained
by other factors.
for the crime example on page 35 how can we prove its a causal relationship
we need to show that : (1) The
reduction of crime actually occurred after the enactment of this measure, (2) the enactment
of the “three strikes and you’re out” measure was empirically associated with a decrease in
crime, and (3) the relationship between the reduction in crime and the “three strikes and
you’re out” policy is not due to the influence of another variable (e.g., the improvement of
overall economic conditions).
how can u find the indep n dep vbles
. The dependent variable is always the property that you are trying to explain; it is
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.
what should the purpose of the research help determine
The purpose of the research should help determine which is the independent variable and
which is the dependent variable.
what is the case w vbles in the real world
In the real world, variables are neither dependent nor
independent; they can be switched around depending on the research problem.
what are some vbles that r never considered dep n why
Some variables,
such as race, age, and ethnicity, because they are primordial characteristics that cannot be
explained by social scientists, are never considered dependent variables in a social science
analysis
what must researchers decide
As researchers, we must decide how to
measure the variables of interest to us, how to select the cases for our research, and what
kind of data collection techniques we will be using
. For instance, suppose we decide to study how
one’s social class is related to attitudes about women in the labor force. Since attitudes
about working women are not directly observable, we need to collect data by…
asking a
group of people questions about their attitudes and opinions. A suitable method of data
collection for this project would be a survey that uses some kind of questionnaire or
interview guide to elicit verbal reports from respondents. The questionnaire could include
numerous questions designed to measure attitudes toward working women, social class, and
other variables relevant to the study.
The type of statistical operation we employ depends on …
how our variables
are measured
The correspondence between the
properties we measure and the numbers representing these properties determines …
the type
of statistical operations we can use
The degree of correspondence also leads to different
ways of measuring—that is, to distinct levels of measurement. In this section, we will
discuss three levels of measuremen
nominal, ordinal and intervaql ratio
At the nominal level of measurement
Because the different categories (e.g., males vs. females) vary in the quality inherent in each
but not in quantity, nominal variables are often called qualitative. Other examples of
nominal-level variables are political party, religion, and race.
nominal level of measurement
Numbers or other symbols are assigned to a set of categories for the purpose of
naming, labeling, or classifying the observations. Nominal categories cannot be rank-ordered.
Nominal variables should include categories that are both exhaustive and mutually
exclusive. Exhaustiveness means…
that there should be enough categories composing the
variables to classify every observation. For example, the common classification of the
variable marital status into the categories “married,” “single,” and “widowed” violates the
requirement of exhaustiveness. As defined, it does not allow us to classify same-sex couples
or heterosexual couples who are not legally married
we can make all vbles exhaustive how? when is this not recommended
by adding the category “other” However, this practice is not
recommended if it leads to the exclusion of categories that have theoretical significance or a
38
substantial number of observations.
mutual exclusiveness
means that there is only one category suitable for each observation. For instance, the categories Protestant and Methodist are not
mutually exclusive because Methodists are also considered Protestant and, therefore, could
be classified into both categories.
ordinal measurement
t Numbers are assigned to rank-ordered categories ranging from low to high.
