Study Guides (248,151)
Canada (121,346)
Sociology (152)
SY280 (9)
Final

SY 280 Exam.doc

30 Pages
239 Views
Unlock Document

Department
Sociology
Course
SY280
Professor
Linda Quirke
Semester
Fall

Description
SY 280 Exam • Ch 6, 11, 16, 7 • Don’t need to do extra readings • 36 MC • 3 fill in the blank • 4 short answer( 2 worth 4, one 6, one 5 ) • 1 essay worth 12 marks • Ethics essay questions on final exam • Validity question on final • Know general information • Structured Observation/Systematic Observation o Social desirability effect o What is Structured Observation  Watch people, record events  If people know they are being studied may act differently o Using Composite Measures  Put together two or more indicators and get composite measure  Multiple indicators • -To measure post partum depression might ask: o Variety of q’s that would indicate  Proxies • Can indirectly measure attitudes, beliefs, feelings • Hard to get at o The researcher formulates explicit rules outlining what behaviours are to be observed and how the observations are to be recorded o Observation Schedule  A clear focus, research problem clearly stated, who is being researched  Mutually exclusive and exhaustive  Simple classification scheme, train observers  Problems if schedule needs too much interpretation o Reliability  Inter observer consistency, how closely they agree when coding  Intra observer consistency, consistency in the using the schedule by a single observer over time o Validity  Measure what it is supposed to measure o Internal  Intersubjective, expert based validity • Based on internal logic of the scale  Look at the items in a scale- ask do the items make sense  Face • On the face of it do these items in this scale make sense  Content • The degree to which indicators cover all of the generally accepted meanings of a variable o External  Criterion • Valid o Predict another variable • Dealing with a second variable (ex driving skill) that occurs in the future o Written driving test score number of car accidents • Predict which drivers are going to be safe versus dangerous drivers? • Can it predict the outcome??  Construct • Ability of a scale to predict some other variable that is theoretically related to the scale • link between scale and other variable est by formal theory • ex: drinking problems spousal abuse o scale shows ppl who drink a lot more likely to abuse their spouses  Errors • Variability between observers or over time makes the measure unreliable and therefore invalid • People will act differently if they know they are being observed o Field Experiment  Study where researcher intervenes in a natural setting to observe the consequences  Don’t know they are being studied o Criticisms  Risk of imposing an inappropriate observation schedule on the setting, not much is known  Less able to get at the intentions behind behaviour, meaning people attach their behaviours  Generate small bits of data • Indexes and Scale o Index (presence or absence of variables)  Accumulating scores assigned to individual attributes  List of questions to determine, Yes/no, presence or absence of something  Combination of two or more indicators  Creates single composite score  Involves traits that are part of the variable  List traits that you associate with ex marriagability, occupational satisfaction o Scale (strength and intensity)  More complex combination of indicators  Measures strength and pattern of responses; some responses weak other strong degree of the presence of a variable  Measure in which the researcher captures • Intensity of feeling, belief, preference, direction, level, strength  Gradient, high, low, agree, disagree, (not yes/no) o Adding scores for several indicators o Level, degree of strength, intensity o Likert Scales  Items involve a variable’s intensity  Intensity involves the level or extent of agreement a person has with items in a scale  Strongly agree, strongly disagree o Bogardus Social Distance Scale  Measure possibility for conflict  Ordered statements, socially intimate to distant  Logic of scale—people who refuse one indicator are unlikely to accept more socially intimate items  Closeness of people with members of diverse social groups  Extent to which people are to have contact with other social groups  Ex: Hierarchy of possibility • Marry, Club Members, Neighbors, Co-worker, Citizens, Visitors, Exclude from country o Semantic Differential  Measure opposite opinions along a continuum  Indentify a number of dimension that can be linked to the variable you want to measure then pick polar opposite terms  Circle number to indicate where you think you fit on a continuum between two opposites  Shy 1 2 3 4 5 Outgoing • Population and Sample o Pop= N o Sample= n • Sampling Terms o Representativeness: sample accurately reflects the population, same proportional data for both groups  Able to generalize o Sampling Bias : those chosen are not typical, not representative, deliberate o Sampling Error: sample not representative of population due to chance o Element  Unit (people, groups, organizations) about which info is collected o Sampling Frame  List of elements that we have  Physical list of units in the population o Response Rate  The % of the population that participates o Sampling Size  Bigger is better, more representative  Reduce error • Probability Sampling o Best kind o Permits inferences about population o 3 types  Simple random • Randomly choose the sample, from the population  Stratified • Dividing the population into subgroups, by criterion and selecting either a simple random sample or a systematic sample from each of the resulting strata  Systematic • Take a sampling interval to get the sample size you want from the population; every n person • Ensure there is no order or pattern in the sampling frame  Multi-stage cluster sampling • The primary sampling unit is not the units of population, but a cluster • Clusters first then subunits within • Non Probaility Sampling o Deliberate selection of specific units of analysis o Cannot infer to the larger population o 3 types  Convenience • Used because the elements are available to the researcher  Snowball • Form of convenience sample • Makes initial contact with a few people then use them to establish contacts • Very unlikely to be representative • Used in qualitative research  Purposive or Judgmental Sampling • Juqgment of expert in selecting classes o Knowledge of population and nature of research o Select typical cases o 3 appropriate situations  Informative unique cases  Members of specialized population • Cant infer to larger population due to judgmental sampling  Particular types of cases • Not representative  Quota Sampling • Produce a sample that reflects a population in terms of the relative proportions of people in different categories, such as those pertaining to gender, ethnicity, age, socioeconomic status and region of residence • Criticisms o Not likely to be representative o Those near the researcher will be chosen o Eligibility may be false • Positives o Cheaper, no travelling, faster, useful for first time testing • Unobtrusive Methods o Act of research= no impact o Obtrusive= lack of validity o Use things the people have produced to get at attitudes and opinions o Typical behaviour is seen, not aware they are being studied • Content analysis o Systematic description and analysis of content of any sort of recorded information o Quantitative  Overt, obvious  Key word search (obesity )  Quantify content; Easier to code info  Replicability  Manifest content • Reliable less valid o Qualitative  Construct meaning of documents and texts  Categories emerge out of the data  Recognize significance of context  Underlying content  Latent content • More valid less reliable o Steps in Content Analysis  Depends on research question 1. Review of literature 2. Define population 3. Sampling 4. define units of analysis 5. create categories for coding 6. coding o Advantages and disadvantages 1. + 1. Transparent 2. Easy replication 3. Longitudinal analysis 4. Does not change behaviours 5. Flexible o Add new categories as they emerge 6. Inexpensive 2. – 1. Documents o Maybe no trail left behind 2. Interpretation 3. Speculative 4. Atheoretical • Historical Research o Idiographic vs nomothetic  Sociology usually doesn’t use Idiographic  Idiographic is Qualitative, exhaust all possible explanations  Nomothetic is Quantitative, partial explanation, most important factors and considerations o Sources of data  Archives, Institutional Records, Extent documents, Don’t have to collect, already there, Not produced for the purpose of social research, Need to be preserved for analysis, Unobtrusive and non reactive o Extant documents: those that are still around today o Criteria to evaluate  Authenticity- real  Credibility- believable  Representativeness- true to time period  Meaning-what was it for? o Artifacts- durable  Durability • Access, Only those that have survived can be looked at  Intentional traces • Left behind on purpose or not • Reliability o Those collecting did it reliable o Ethics o Validity and reliability o Example  Jesuit Relations  1600’s-1901  Trace the steps • Handwriting letters to the publication in North America • Sent from missions to superior in Quebec • Printed in France • 1858 published in Canada • Jesuit relations and allied documents • Being able to evaluate their value for research • Letters written by Priests about their day to day life collected over 150 years. Complied and edited by the Superior then sent to Paris th head Jesuits, may have been altered?? Published in mid 17 century. Canada released in 1858  1600: from Handwriting to Publication • Letters sent from missions to the superior in Quebec • Priest write about day to day life • 1611-1768 • Compiled , edited by superior, then sent to head Jesuit Paris office • Message might get altered in some way during transfer • 1632-1673 o Reports published in France o Happenings in New France o Served to publicize what was happening  Missions  Relations: Lost and Found • Letter written 1610=1791 o Annual reports • After 1673: obscurity • 1858: Canadian government reissued Cramoisy series • 1896-1901: “’Jesuit Relations and Allied Documents o 73 volumes  Republication, 1890’s • Reuben Gold Thwaites • Wanted the relations in print  How can we use Historical Documents • Reasons to be cautious o Accuracy of records  Replication and corroboration will increase o Bias in data sources o Are these data all bad? Better than nothing? o Example 2  Census 1861  Organizational fiasco  How bad was it  Enumerators work • Teachers (women = big problem)  Schedules • Printing problems • Most didn’t have enough sheets o Use other paper, Squeeze extra names on • Size of the paper o Massive, Had to write on the floor • Delay o Not enough enumerators to cover everyone • Weather o Done in the winter time o Ask the same family about other families they know to not have to go outside • Power issue o Man would answer all the questions of the house; innaccurate • Gender bias o Male enumerators might change info that they were given o Women in unusually position may be recorded as lesser occupations  Commissioners corrections • Interpretation • Re-enumerate o Gaps in the data o Bad enumerator o Go back out the house to collect data again o Issue of validity for collecting data much after • Errors o Filling in inaccurate or missing data based on their knowledge • Incomplete returns o Fill in what they thought was true   Conclusion: unreliable data • Ethics o Milgram Experiment  person giving shock is the participant  told they must continue  What extent will people follow orders  Extreme psychological stress, falsely informed, not informed consent o Tea Room Trade  Data collection: St Louise 1965  Study secret gay sex in public washrooms  2 stages • Ethnography o Public washroom o Direct observation, don’t know they are studied o Took down license plates, o Built trust by pretending he also did so • Administering interview o Asked men back based on license plate o Asked questions about their health etc  7 issues to consider: ethical treatment of human participants • Voluntary participation o Motivation? Payment, extra grades, cant entice too much o Wont get rep sample • No harm o Informed consent, psychological harm • Anonymity o Cant put response to a person • Confidentiality o Promise not to identify, pseudonyms • Deception o Must be justified, • Debriefing o Explain true purpose o Erase negative effects • Analysis and Reporting o must report all results, make shortcomings known, o Possible source of harm  Participants characterized negatively  “Im not very fond of my skeleton image” || Class Notes November 4, 2010 • Indexes and Scales o Index  Accumulating scores assigned to individual attributes  List of questions to determine  Adding scores form several indicators  Yes/no, presence or absence of something  Combination of two or more indicators  Measure in which a researcher adds or combines several distinct indicators of a construct the measure in which the researcher condenses down into a single score= composite score  Involves traits that are part of the variable • Link traits back to variable  Ex.1: Occupations • Does it pay a good salary (yes/no) • Is it secure from layoff s or unemployment • Is the work interesting and challenging • cxvAre working conditions ood • Are there opportunities for career advancement and promotion • Total yes or no responses, greater # of yes responses is the highest score  Ex.2: Marrigeability • Extent to which one is viewed as a candidate for marriag e o Come up with traits that someone could have that would make them marriageable • attractive • employed • intelligent • personality • these traits can all be associated with HIGH marriageability  ex.3: Status of Women • indexes allow us to compare different countries o Scale  More complex combination of indicators  Measures strength and pattern of responses; some responses weak other strong degree of the presence of a variable  Measure in which the researcher captures • Intensity of feeling, belief, preference, direction, level, strength  Gradient, high, low, agree, disagree, not yes/no  Arranges responses or observations  Measure of a variable expresses a numeric score  One or multiple indicators  Likert Scales • Items involve a variable’s intensity o Intensity involves the level or extent of agreement a person has with items in a scale o Strongly agree, strongly disagree  Ex. 1: Relationship Satisfaction Likert Scale • The extent to which one feels that one’s relationship is satisfying- how do we measure this? • One indicator- I am satisfied with my relationship o Response—do you agree with this statement?? o Using them  Give info about variables, possible to increase quality of assessment  Helpful for data reduction  Condense and simplify the info  Summarize several indicators into a score • Creating Indexes with Likert Scales o Do you think Laurier is a good school?  Ask one thing its just a scale  Combine multiple questions and make an index o Would you advise a friend? o Do you think that WLU has prepared for life after graduation o Total all scores from scales to create an index • Bogardus Social Distance Scale o Measure possibility for conflict o Ordered statements, socially intimate to distant o Logic of scale—people who refuse one indicator are unlikely to accept more socially intimate items o Extent to which people are to have contact with other social groups o Ex: Hierarchy of possibility  Marry  Club Members  Neighbors  Co-workers  Citizens  Visitors  Exclude from country • Semantic Differential o Measure opposite opinions along a continuum o Indentify a number of dimension that can be linked to the variable you want to measure then pick polar opposite terms o Circle number to indicate where you think you fit on a continuum between two opposites o Shy 1 2 3 4 5 Outgoing o Passive vs dominant Novemebr 9, 2010 Sampling • Population and Sample o Population  Group we cant to know about all units from which the sample is to be selected o Sample  Segment of the population selected, subset population; group studied directly  Who we collect the data from  Use to generalize findings o Ex:  Population N=2 million (capital N for the total population)  Sample n=500( lower n for the sample, different cases) • Representativeness and Generalization o Representativeness= sample accurately reflects the population  When the traits that we find in the population appear I the same proportion in the sample o Generalization  Matching up sample and population, confident in generalizing results to larger population • Sampling Bias and Error o Bias  Individuals selected for study are not typical, not representative of larger population • Deliberate exclusion of including certain cases • Faulty strategy was used to select the sample • Convenience o Error  Sample not representative of population due to chance or random factors  Accidental  Easier to fix • Less serious than sampling bias • Elements and Sampling Frames o Element  Unit (people, groups, organizations) about which info is collected o Sampling Frame  List of elements that we have  Physical list of units in the population • Probability Sampling o Allows tests of statistical significance, permits inferences about population • 3 types o Simple random sample  Each unit has equal chance of inclusion  Most basic  Define population (N=1500)  Comprehensive sampling frame  Decide sample size ( n=150) o Stratified random sample  Divide population into relevant sub populations • Sample randomly from each sub-population • Ensure proper proportions o Researching students from different programs o Systematic sampling  Every k element in total list is chosen th th  List of all elements, choose every 12 person, or every 50 organization in list • Sampling interval o Standard distance between elements selected I the sample o Population/ (divided) sample size • Random start  How many people in population and sample • Non- Probability Sampling o Deliberate selection of specific units of analysis November 11, 2010 • Snowball Sampling o Begin with one case o Members of difficult to reach populations o They will know other people in the same situation o Build sample  Interrelationships o Minorities o Unpopular beliefs o Members of a controversial group o Helps build trust • Purposive or Judgmental Sampling o Judgment of expert in selecting classes  Knowledge of population and nature of research  Select typical cases  3 appropriate situations • Informative unique cases • Members of specialized population o Cant infer to larger population due to judgmental sampling • Particular types of cases o Not representativeness • Sample Size and Non- Response o Sample size: bigger = better o Decreases sampling error o More precise and representative o Depends on time, most importantly $ o Response Rate  Percentage of sample that agrees to participate in a study • Unobtrusive Research and Content Analysis o Unobtrusive Methods  Do not make research participants aware of their being studied  No Reactivity • Typical behaviour is seen  Studying without affecting  Act of research = no impact • Gives more validity • No false responses  Use things that they have produced to get at attitudes and opinions  Validity question on final o Obtrusive  Lack of validity o Content Analysis  Systematic description and analysis of content of any sort of recorded information  Examine • Documents, images, searches for specific categories of
More Less

Related notes for SY280

Log In


OR

Join OneClass

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

Sign up

Join to view


OR

By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.


Submit