Class Notes (837,550)
Sociology (2,430)
SOCB05H3 (99)
Lecture

# LECTURE NOTES SEPTEMBER 27 2012

4 Pages
54 Views

Department
Sociology
Course
SOCB05H3
Professor
Rania Salem
Semester
Fall

Description
SEPTEMBER 27 2012 (LECTURE):  TA [email protected]  Composite measures a series of items which try to do 2 things. o Try to capture different dimensions of concepts o Capture the different manifestations of a dimension  2 different types of composite measures: o Indexes  consists of survey questions. Points for the questions are accumulated and assigned to different items. Points at the end, result in the assign responses o Scales  assign scores to patters of responses, they usually recognize some responses reflect the concept weakly and others strongly  STEP 1 IN CONSTRUCTING AN INDEX o FACE VALIDITY  common sense of correspondence of the item and the view on the topic o UNI-DIMENSIONALITY items that don’t have to do with the topic o VARIANCE certain item against that topic wont help the survey because it will result in zero responses to that particularly question.  STEP 2 EXAMINE THE EMPIRICAL RELATIONSHIP o Answer to one item in the survey, should help us see how the other questions will be answered  If this isn’t possible, the index has a problem  STEP 3 ASSIGNING SCORES TO THE INDEX ITEMS  STEP 4 INDEX VALIDATION o Internal (item analysis) taking out extra questions which might be similar but at the same time still being able to have a proper index o External relate aspects of the measures. Having “abortion behaviour” in an “abortion” survey works. It wont work with “homo”  In scale, the items are ranked. Opposite of index  Scaling method: lickert scaling o Check what percentage of the responses for an items would get a yes in the index o Strongly approved to strongly disagreeing  probability sampling: key terms: o Population  the theoretically specified elements of a study o Sample  the elements of a population actually selected for observation in a study o Representative sample  a sample that contains the same distribution of characteristics that exist in the population o Sampling frame  list of elements from which a probability sample is selected  Defining two main sampling techniques in survey research o Sampling  process of selecting items o Probability sampling  sampling based on a process of random selection that gives each case in the population an equal or known chance of being included in the sample o Non-probability sampling  sampling based on a process other than random selection  Probability sampling designs multistage cluster sampling o Within each cluster, elements can be selected using random, systematic, or stratified sampling o How should clusters be sampled> probability proportionate to size  Non-probability sampling designs: reliance on available subjects o Select a number of cases from those what are conveniently available  Non-probability sampling designs: purposive sampling o Use judgement to select units that are representative or typical of the population  Non-probability sampling design: snowball sampling o Collect data on a few members of the target population that are easily located o Ask those cases to refer you to other members of target population; and so forth  EXAMPLE prostution making the law for them better for their own safety o Sampling  non- probability because its hard to find them and there is no directory for them o Method snowball o Population  prostitutes o Sample  narrow it down to the location you are in. 20 Toronto SEPTEMBER 27 2012 (CHAPTER 6 & 7):  The terms index and scale are typically used imprecisely and interchangeably in social research literature  Both scales and indexes are ordinal measures of variables such as religiosity, alienation, socioeconomic status, prejudice, or intellectual sophistication  Both scales and indexes are composite measures of variables o That is measurements based on more than one data item  An index is constructed by simply accumulated scores assigned to individual indicators  A scale, is constructed by assigning scores to patterns of responses, recognizing that some items reflect a relatively weak degree of the variable while others reflect something stronger o Scales are generally superior to indexes, because scales take into consideration the intensity with which different items reflect the variable being measured  4 main steps in the construction of index 
More Less

Related notes for SOCB05H3
Me

OR

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Join to view

OR

By registering, I agree to the Terms and Privacy Policies
Just a few more details

So we can recommend you notes for your school.