Home > Preview
The flashcards below were created by user
on FreezingBlue Flashcards.
Purposes of research (3)
- 1. Exploration
- 2. Description
- 3. Explaination
• Why do people research for exploration?
- • To provide a beginning familarity with a topic
- • Valuable to social research -- breaking ground
- • Yields grounded theory
- • Difficulty issue of representativesness
• Why do people research for description?
- • Describe situations or events
- • Can be quatitative (US census, Gallop poll) or qualitative (in depth study of what its like to walk in the shoes of people described)
• Why do people research for explaination?
- • Answers WHY questions
- • Typically deductive in nature - rigorously testing a hypothesis or predicted relationship between variables
• What is ecological fallacy?
- • Making assumptions about individuals as units of analysis based upon your examination of groups
- • E.G. Studying voting behavior by precincts and assuming we know individual characteristics
• What is reductionism?
- • Overly strict limitations on kinds of concepts and variables considered as causes
- • E.g. sociological or psychological
• What's unit analysis?
- • Units that we initially describe for the purpose of aggregating their characteristics in order to describe some larger group, or explain some abstract phenomenon
- • Who or what is studied
• Types of units of analysis
- • 1) Individuals
- • 2) Groups
- • 3) Social artifacts
- • 4) Social interactions
• What are individuals?
Most typical norm of general understanding. May be described in terms of being members of a group, but individual behavior is what we're interested in
• What are groups?
- • Not the same as studying the inviduals within the group.
- • E.g. gangs, families, social organizations
• What are social artifacts?
- • Studying the products of social beings or behavior
- • E.g. books, buildings, jokes, poems, historical documents
• What are social interactions?
- • Studying the social event
- • E.g. weddings, court cases, race riots, traffic accidents
• Different types of variables.
- • Independent
- • Dependent
- • Control
• Independent variable
a variable that explains or causes something
• Dependent variable
the variable we measure the change on, it "depends" on the independent variable, the effect
• Control variable
when we seek to control for an extraneous variable
• Extraneous variable
represent alternative explainations for the relationships between individual and dependent variables
• Difference between variables and attributes
Variables are logical groups of attributes, and attributes are characteristics or qualities that describe an object. people are carriers of variables
e.g. variables gender has 2 attributes: male and female
-theories describe logical relationships among variables
• How is independent variable related to dependent variable?
bc any change in one affects the other
e.g. the more hours a student studies the better the student will do on the exam
• Four levels of measurement
- • Nominal
- • Ordinal
- • Interval
- • Ratio
• What's nominal measurement?
- • It has only the properties of exhaustiveness and mutal exclusiveness
- • E.g. gender, religion
• What's ordinal measurement?
- • In addition to exhaustiveness and mutal exclusiveness, attributes are rank ordered
- • E.g. social class, client satisfaction
• What's interval measurement?
- • In addition to exhaustiveness, mutal exclusiveness and ability to rank order, the logical distance between attributes can be expressed in meaningful standard intervals. Only common examples are constructed measures
- • Eg IQ scales, where differences between 1-- and 110 same as between 120-130
• What's ratio measurement?
- • In addition to all the structural qualities of 3 preceding levels, it's also based on a true zero point
- • E.g. age, number of services
the process through which we specify precisley what we mean when we use a particular term, the precise working agreements about meanings. It involves specifying dimensions and the indicators for each of those dimensions.
Concepts can not be measured directly, they're mental images
E.g. prejudice, compassion, self esteem
• What's reliability?
Whether a particular measure/technique applied repeatedly to the same object yields the same results each time - has to do with random error in measurement
• How to create reliable measures
- • Ask only questions about things respondents are likely to know
- • Use measures previously proven to be reliable
- • Follow up verification of sub sample of survey data
- • Clarity, specificity and training
• Types of reliability
- • Inter-rater reliability
- • Test-retest reliability
- • Parallel forms reliability
- • Internal consistency reliablity
• What's inter-rater reliability
- • if we need to assess degree of agreement or consistency between observers
- • -widely used in clinical measures
- • -70% or above (correlation) good level of agreement
• What's test-retest reliability
- • Determines how stable (consistent) a measure is over time
- • -administer the same instrument under same conditions on 2 seperate occasions and correlate
• What's parallel forms reliability?
constructing a 2nd instrument that is the same ast the 1st instrument (RARE!)
• What's internal consistency reliability?
- • Assessing the homogeneity of the measure
- • -divide the instrument into equal halves
- • -assess the correlation of total scores of each half
- • -coefficient alpha the average (mean) of all correlations
the extent to which a measure is measuring the meaning of the concept
• Types of subjective validity
- • 1. Face validity: does the measure appear to be measuring what the researcher is trying to measure?
- • --eg comaring shoe size with IQ
- • 2. Content validity: refers to the degree to which a measure covers the range of meanings included within a concept
• Refer to wheel of science in notebook
• Qualitative (inductive)
emphasize the depth of understanding associated with idographic concerns, tap the deeper meanings of particular human experiences and are intended to generate theoretically richer observations that aren't easily reduced to numbers
• Quantitative (deductive)
emphasize the production of precise and gerneralizable statistical findings and are gnerally more appropiate to nomothetic aims. Used most often when we want to verify whether a cause produces an effect in general