Ch. 1 Why the Social Researcher Uses Statistics?

CH. 1 Why the Social Researcher Uses Statistics?

The Nature of Social Research
• Social scientists attempt to explain and predict human behavior
– Social scientists uses scientific methods to study the social world: human behaviors & interactions
– Variables: characteristics that differ or vary from one individual to another or from one point in time to another
• Ex. Age, social class, and attitude or unemployment, crime rate and population
– Constants: characteristics that do not vary
• Ex. The gender of the person who gave birth to you is female. Thus, in any group of individuals, gender of mother is the constants ‘female’ and does not change
– Unit of observation- what are we measuring?
• Measuring the individual & comparing individuals
– Aggregates – how measures vary across entire collections of people
– Hypotheses- usually a statement about the relationship between two or more variables
• Independent variable- presumed cause; causing a change in the other
• Dependent variable- presumed effect; being influence, affected or changed
• Ex. Education vs. income; the better the education(independent variable) the greater the income (dependent variable)
Why Test Hypotheses?
• Common sense vs. social science research
– Common sense: observation are generally based on narrow, often biased preconceptions & personal experiences; can be misled by assumptions & stereotypes
– Social science research: scientific collection & analysis of data; based on empirical data because we are collecting data and doing research
• Ex. Mass Murderers
• Popular thinking (medial portrayals)- mass murders are insane, individuals who go berserk or running crazily, expression their anger & showing aggression; total strangers to their victims; wrong place, wrong time such as shopping mall, trains, fast food restaurant
• FBI reports: according to reports; mass murderers are not crazy, plan their attacks & murders; targets spouse & all their children, or bosses and their co-workers…most mass murders occurs within families or among acquaintances.

The 5 Stages of Social Research
1. Create testable hypothesis based on relationship you want to study
2. Develop appropriate set of instruments (ex. surveys, questionnaires or interview)
3. Collect data
4. Analyze data
5. Interpret and share results of analysis
Levels of Measurement
- Classify or categorize at the nominal level of measurements
- Rank or order at the ordinal level of measurements and
- Assign a score at the interval/ration level of measurements
• Nominal level- classify or categorize data (least sophisticated)
• Naming or labeling
– Counting frequency of each category
• Etc. dividing a class by sex and counting males from females; categorize by race & religion etc.
– Categories must be:
• Mutually exclusive- can only fit into one
• Exhaustive- everyone must fit into a category
– Not graded, ranked, ordered or scaled for qualities
• Ordinal level- rank or order data (more sophisticated)
– Puts data in order, BUT does not tell us the magnitude of difference between numbers
– Not possible to assign scores to cases as located at points along the scale
• Etc. social class: lower class, middle class & upper class etc. (cannot quantify how much income an individual make etc.)
• Interval level- scores data so that the exact distance between them is known. (most sophisticated)
– not only indicate the ordering of categories but also the exact distance (can quantify)
– Use constant units of measurement that yield equal intervals between points on the scale
• Ex. Ratio level (how many lbs. you weigh, siblings you have, how long it takes students to complete an exam) or unit of measurements ($, cents, Fahrenheit, Celsius yard, feet, ,minutes & seconds)
• Ex. Ration of how we scale etc.
– Treating some ordinal variables as interval
• Sometimes we treat ordinal variables as interval when their ordered categories are fairly evenly spaced
• Ex. attitude scales

The Functions of Statistics
- Descriptive statistics: arriving to a conclusion or obtain results
• Ex. characterized the distribution of ages of everyone in the class
• Statistics: sets of techniques for reducing quantitative data to smaller, more manageable amount of data
- Decision making (aka Inferential Statistics): Testing hypotheses in order to make inferences.
• When testing hypotheses its often necessary to make inferences
• Make decisions based on data collected on only a small portion (sample) of the larger group we want to study (population)
- Can we generalize our findings from our sample to populations? (drawing inferences)
• Every time social researchers test hypotheses on a sample, they must decide whether it is indeed accurate to generalize the findings to the entire population from which they were drawn
• Statistical tools can help us test for significant difference
• Statistics: set of decisions making techniques that aid researchers in drawing inferences from samples to population and hence, in testing hypotheses regarding the nature of social reality.
Rounding
- Round the final answer to two more decimal digits than contained in the original scores
• Raw scores have 1 decimal place (1.3), so..
• Round answer to 3 decimal places (2.345)
- Don’t round until your final answer (don’t round during intermediate steps)