Content Analysis - "unobtrusive data"  Data created by a bureaucratic system, e. g. police records, or often by the media.  Television or Newspapers either because that is our interest, the media, or as a way of getting information, e.g., on crime reported in the news.

Similar to survey research, except that you do coding instead of interviewing.  Coding means that you assign numbers to phenomena that you observe.  Counting things.  Each of your variables is coded from the published information.

Conceptualization.
Measurement.  Reliability and Validity.

The Review Glossary is not adequate as a guide to this chapter.  Some points to be covered:

• Field research is needed to gather data on processes, especially those involving personal interaction
• It is good for observing behavior rather than getting reports of behavior or attitudes
• It takes a lot of time, and is less likely to be replicated, leading to problems of reliability.  Would another researcher observe the same thing?  Validity issues are difficult to assess, what abstract concepts are being used?
• By its nature field research is likely to be inductive, people go to see what's happening rather than to test a theory.  But observers may have a "hidden agenda" that leads them to see what they want to see.
• Selection of sites and topics is likely to reflect the special interest of the researcher.  There may be a bias toward the bizarre, or towards cases that will prove a particular point.
• Getting access is problematic, groups may not want to be observed.  Sometimes deception is used, which raises ethical issues.  The whole question of ethical review is difficult to apply to field research, e.g., "informed consent."
• There can be structured and unstructured observation.  You can count things.  Observation through a one-way mirror is sometimes possible, e.g., of day care centers.  This might be done with WEB-cams.
• Notes are important, need to be recorded quickly.  Data base programs can be used, here there might be validity issues, in the coding of observations, e.g., was a certain remark a case of "sexual harassment"?
• Accounts tend to be descriptive, there is a bias toward things that are interesting.  It is possible to combine field work with quantitative data collection.

Some example of field research:
Margaret Mead, the only anthropologist (or sociologist) to get her own postage stamp, won fame through field work, primarily her book Coming of Age in Samoa.  Later, this book was denounced by anthropologist Derek Freeman in his book Margaret Mead and the Heretic : The Making and Unmaking of an Anthropological Myth.Anthropologists have come to Mead's defense, and have restudied the case, but I would have to agree with your text that "had Mead come back from Samoa with an accurate ethnographic report, it would not have made her famous."  Here is the NY Times Review of Freeman's critique of Mead.
     More recently, there has been a raging controversy about the book Darkness in El Dorado about research on the Yanomamo in Venezuela is the latest ethical controversy, which also raises important methodological questions.  Many of the book's allegations, however, have been contested by the National Academy of Sciences.
  The combining of fiction with factual research is increasingly common both in anthropology and in biographies.  Sometimes this is openly done as a literary form, in other cases such as that of Rigoberta Menchu, it is only admitted when critics discover it.   The Rigoberta Menchu Controversy by Arturo Arias. 
There are many problems with field research:  ethical issues, problems of reliability and validity when data are gathered by only one researcher, etc. A controversial book is Laud Humphrey's Tea Room Trade, which raises ethical issues. He studied gay sex in a men's room in a park in St. Louis, without informing the participants what he was doing.
    Field researchers sometimes seem to find examples that fit their preconceptions, and their work is often ignored by those who do not like the results, e.g., Leon Dash's book When Children Want Children and Rosa Lee which are just ignored by welfare advocates who prefer more sympathetic treatments.  One of the best field studies is Kathryn Edin's book Making Ends Meet. which is highly sympathetic to the mothers.  However, Edin collected statistical data as well her illustrative observations.  The statistics showed that almost none of the mothers actually lived off their grants alone.  Eli Anderson's book Streetwise on men in a Philadelphia ghetto has been well received, in large part because goes beyond one-sided advocacy.
    A great strength of field work is observing behaviors that the people themselves don't understand or aren't even aware of., or at any event, are unable or unwilling to talk about.  Anthropologist Jules Henry spent a week living in each of the homes of several children who had grown up mentally ill, trying to discern patterns in the family interactions that contributed to the illness.   Myra Bluebond-Langner's book The Private Worlds of Dying Children has been very influential;  she has just published a sequel called In the Shadow of Illness : Parents and Siblings of the Chronically Ill Child    Field reserch offers a richness of description and possibility of new insights that is unparalled by any other method.  Unless it is supplemented with other methods, it does not provide statistical data, and it is hard to replicate.
    Myra Bluebond-Langner of our Anthropology Department wrote a classic, The Private Worlds of Dying Children, and more recently, In The Shadow of Illness.

November 7 -    Levels of meaurement:
    Religious affiliation is measured on what level of measurement?  a,  nominal  b. ordinal  c.  interval  d. ratio
     Income in dollars is measured on what level of measurement?   a,  nominal  b. ordinal  c.  interval  d. ratio
      Income in categories is measured on what level of measurement?   a,  nominal  b. ordinal  c.  interval  d. ratio

  The chisquare statistic requires which level of measurement?   a,  nominal  b. ordinal  c.  interval  d. ratio
  to do a chisquare, all you have to do is put things into categories and compute a cross-tabulation
                                                      Rich  Poor
                             Methodist
                             Presbyterian
                             LDS
                             Catholic

  The regression or correlation statistic requires which level of measurement?   a,  nominal  b. ordinal  c.  interval  d. ratio

                     We do a scattergram and we need interval data.

  Which statistic is based on the difference between the expected and observed frequencies:   a.  correlation  b.  cronbach's alpha  c.  the mean  d.  the standard deviation  e.  chi-square

   Which statistic is a measure of central tendency:   a.  correlation  b.  cronbach's alpha  c.  the mean  d.  the standard deviation  e.  chi-square

Which statistic is a measure of dispersion?   a.  correlation  b.  cronbach's alpha  c.  the mean  d.  the standard deviation  e.  chi-square

THree criteria for a causal relationship;  time sequence, correlation between the variables and non-spuriousness, not determined by any other variable.

What are the two types of test variables?  a.  antecedent  b.  intervening

November 3 and 5 - we went through exercise 5a in the workbook, controlling for variables
October 31 - Interviewing begins.  I ran off some questionnaires, but then Jon'a had to make corrections, so there are a few minor points that need changed on the once I'll hand out today:    Final Version is Here.

Here is her note about changes::

So, during our practicing today, we found an error (in the set for 69, choice D should be "very adequate"), so I fixed that and added the respondent ID number to the forms.

I also added skip directions to question 12 choices D and E and italicized the directions for the school/work questions.

Finally, I added more text to 25, so the direction of the scale is clearer.

You could theoretically use the old forms, since the changes are mainly cosmetic.
 

October 28 - Causal Analysis - Chapter 5.

The Art and Science of Cause and Effect. (powerpoint)

Probabilistic cause, not an absolute cause, not a cause that is sufficient or necessary.   "Cigarette smoking causes
cancer."  WHat we mean is, smoking cigarettes increases the likelihood of getting cancer.  How much?

There are multiple causes for everything.  What we want to find out is how much each thing contributes.  There are also
causal linkages, or indirect causes.  A causes B and then B causes C.

Diagraming causal models.  We put the dependent variable at the right.  We draw arrows going into it for each causal
variable that effects it directly.  Then we can have arrows that go into the arrows, steps into the causal analysis, as in
this sample file:
http://crab.rutgers.edu/~goertzel/homomale.htm

Criteria of Causation - how do we know that something is a cause of something else.

1.  Time Order.  The cause comes before the effect.  Sometimes we sort out the time order theoretically, we assume that
education preceeds employment.  Or we can use a research design that involves gathering data at two points in time.  If
you don't have measurements at two points in time, this is shaky.

2.  Correlation.  The two variables vary together.  When one is high, the other is high OR when one is low the other is
high.  This gets at the degree of causation, the higher the correlation the strong the causal relationship.

3.  non-spuriousness,  we want to know that the correlation is not cause by something else.  We can test this with an
experimental design, if feasible.  Or we can use statistical controls, which are not quite as convincing but its all you do
in many cases.

We test for non-spuriousness by introducing controls.

Causal Models:  representations of the complex causal relationships between variables.  Variables have different causal roles, but this is determined by our causal our causal model, it is not inherent in the variables.   One person's cause can be another's effect.

Dependent Variable - that is what we want to explain.  Often these are opinions or behaviors

Independent Variable - what we use to explain it.  Often there are traits or physical characteristics, e.g., sex or race,
almost always independent.

If you study the relationship of race on voting, for example, race would be independent and voting dependent.

Antecedent variables, things come before the independent variable.  This helps us to deal with a causal chain.
Antecedent variable cause IV which causes the DV.
If the antecedent variable "explains" the relationship, we have an "explanation", we say it is "spurious".

Intervening Variables, this that are intervening, e.g.   Race determines ideology which determines the vote.
This is an "interpretation" it tells WHY the causal relationship exists.
Path Models:  a way of graphically expressing complex causal models.

Example:  Determinants of Adult Homosexuality in White Males.

Example:  The Seattle Social Development Project. 
 

Example:  Manuscript Pages 6 to 9 of Performance Measures of Effectiveness.
 

October 26 - More on sampling methods.
Population/sample.  The population is whatever group we want to study, the sample is the subgroup selected to represent the population.
Are we interested in the population as a whole, or in sub-groups.  If we are interested in sub-groups, we would like to have an equal number of subjects from each sub-group in the sample, regardless of their number in the population.

Stratification:  We are interested in learning about sub-groups, so we sample each of those groups.  Black, White, Hispanic Asian.   Republicans, Democrats, Independents.  Income strata.  Prtobable voters.  Users of binoculars.    You need a list with a measure of the characteristic you want to use.  In practice, you don't have a measure.

Clustering:  When there is no list or when it is impractical to include everybody in the "sampling frame" (list of people you draw the sample from) so you select clusters and sample them, then sample individuals within the clusters.  Typically "census tracts" are used.  Sometimes zip codes, but these are too large.  The only advantage is saving effort and money.

Margin of error.    If you are given a margin of error and asked how large a sample you need, use  n = 1/ m^{2  If they have a 3% margin of error, that is .03 when expressed as a proportion.   46% for Street.  Sample statistic is 46%.  We are 95% sure that the population parameter (the true figure for the population) is between  43% and 49%.  Confidence interval is the sample statistic plus and minus the margin of error.  Katz was 41%}

^{Practical problems:  non-randomness in the sample.  Nonresponse error.  People without phones.  Behavior vs. opinion, predicting how likely it is that someone will turn out and vote.  Samples are weighted to adjust for the difference between demographic traits in the sample and in the population.  The times is biased in its weighting of the sample, increasing the number of democratcs and the number of liberals.}

^{Another source of bias can be the wording of the questions, or even the order of the questions.  When opinion is not reall fixed, you can sway the results with the wording.  Look for behavior vs. attitudes.}

Controversy over New York Times sampling methods raised by Dick Morris's book Off With Their Heads.  He argues that  ''in every single survey'' from Jan. 1, 2002, to the election that year, ''the number of Democrats interviewed is weighted up . . . while the number of Republicans interviewed is weighted down.'' (See NY Times review by Michael Janeway, August 31, 2003).  Timothy Noah in Slate exonerates the Times.

How large a sample do we need?  We have 1000 people and I have 55 and Jon'a has 40.  95 students.  We are going to give each student eight numbers.  Ask you to call all eight.  My hope is we will get at least five per student.
 
 

Oct 24 -  pages 166 on in the workbook.   Introduction to sampling, chapter 4 in the text.
Sample statistic - what the sample says
population parameter - what the real figure is
Even if the sampling is done well, the response rate is less than 100%.
Weighting is done to make the sample more like the population.

   m = 1/sqrt(n)    Solve for N:      m² =  1/n      n * m² = 1     n = 1/ m²    If we need a margin of error of 3%, or .03.   n = 1/ .03²

  ^{If you have a sample size and need to know the margin of error, use}    m = 1/sqrt(n)

   If you are given a margin of error and asked how large a sample you need, use  n = 1/ m²

          In these formulas n = the size of the sample (not the population).    m = the margin of error expressed as a proportion, not as a percent.  Thus, if the questions says "we need a margin of error of 5%, then m = .05.    If we need a margin of error of 5% for three groups, then the answer is  3 * ( n = 1/ m² ) .
 

Oct 22 - we went through pages 155-166 in the workbook.
October 20 - Discussion of Survey Research.  Review of First Draft of Survey for this Semester.
Simple vs. proportional facts.  Qualitative vs. quantitative.  A percentage or a frequency.  Proportional facts can be viewed as trends.  They are more or less at different periods of time, or comparing groups.
We are interested in a population.  Define the population we are talking about, e.g. adults in New Jersey or probable voters.  People convicted of felonies.  People who graduated from our department with a major in CJ or sociology.  Select a sample of the population.  A sub-group that is used to generalize to the population.  If you try to study the whole population, you often fail to do a good job.
Size of the sample.  How big of a sample do I need? Size of the sample does not depend on the size of the population.
How do we select the sample size?  Decide on the margin of error you will tolerate?  Margin of error is equal to one divided by the square root of the sample size.  Sample of 400, the square root is 20.  1/20 = .05 or 5%.  If you interviewed 400, 300 were white, 50 were black and 50 were others.  For the blacks, with a sample of 50, we would have a 14% margin of error.  For the whites, with a sample of 300, we would have a 5.8% margin or error.

Take 300, the square root of 300 is = 17.32     1 /17.32 = .0577  * 100 = 5.8%

What questions to ask.  Key to survey research is that everybody gets asked the same questions, at least if appropriate.  Open ended vs. closed ended.  multiple choice vs. short answer or essay.  Questions have to be clear and unambiguous.
 

October 17 - we did a Descriptive Statistics example in class.
October 15 -    Nominal = ordinal - interval  this are inclusive.   If you have interval measurement, you also can treat it as ordinal or nominal.  Age could be treated as nominal even though inherently it is ratio.    Often we just sort things into Categorical vs. Continous.  Continuous include interval and ratio.  Categorical is nominal and perhaps also ordinal.  Categorical measures use cross-tabulation.  With interval measures we use the descriptive statistics we are going to talk about today.
Today we will discuss Descriptive Statistiscs.  There is a handout on this since it is not in our book.  The handout is also on the WEB site.  Some illustrative data is also available in an Excel File on height and weight.

Standard deviation  - about two thirds of the cases will be within one standard deviation of the mean.  About 95% of the cases will be within two standard deviations of the mean.
 

October 13. Trochim on scaling.  Note:  your book distinguishes between an "index" and a "scale" but many other books use the term "scale" for both.  "Likert" or "summated" scaling is what your book calles "index construction."  Trochim calls an index a "response format".  Most measurement in sociology and crminal justice uses indexes, but there are several well known ways of constructing scales and indices.  Several are named after their inventors.

1. Thurstone scaling - you have judges rate the items
2. Guttman scaling - you select items on a continuum and test to see whether your respondents answer them consistently in the sense that people who agree with a more extreme item also agree with the less extreme ones expressing the same value.  A well known example is the  Bogardus Social Distance Scale.
3. Likert or summated "scaling"  or index-construction.  You do not try to order or rate the items, you just add up the scores on all of them.
4. Osgood's Semantic Differential is another approach, this involves getting at the emotional resonance of a term by asking respondents to associate it with a series of polar opposite terms.
5. There is also "multidimensional" scaling, but this is not widely used for practical purposes - it is simpler to just have a  bunch of "unidimensional" scales.

There are some current issues involving measurement and scaling, especially with regard to the issue of intelligence.  See: Roy Freedle's article on the  SAT.  The College Board's Response.  The SAT is what our text calls an "index" - the scores is based on adding up the number of items correct.  Freedle is proposing converting it to a "scale" - a test where the score depends on which items an individual gets correct.  He argues that African-Americans tend to do better on harder items than they do on easier ones, and grading it as a "scale" would take this into account.

October 10

Questions on page 67 of the workbook.
1.  We have a trait or concept on which the individuals have different values or go into different categories.  We assign numbers to the values, but sometimes the numbers are arbitrary.
2.  Levels of measurement. discussed below.  First is categories, nominal, then they are in order, ordinal, then we know the distance between them, interval.  Finally, we also have a meaningful zero.

What is a "unit of analysis".  It what we are talking about, it is the unit that has a value on the variable.   Usually they are people.  Or families.  Could also be states or countries, or counties.

"Ecological Fallacy" -  take aggregate data and ascribe it to individuals within the aggregate.

Qaulity of Measures -
   Reliability -  you get the same thing over and over.  Consistency.
inter-rater - two different raters get the same answer.
test-retest, if you take it twice the answers are the same.
internal consistency - are theitems on a test consistent.  Chronbach's alpha is a statistic that measure inter-item reliability.
   Validity  is it "really" measuring what it is supposed to measure.
           Face Validity - does it look right?
            Predictive or criterion validity - does it predict what we want to predict, some "true" measure.  SAT test predicts college or law or medical school grades.
              Construct validity - does the measure perform as our theory says it should.  We use this when we have no criterion.
             convergent validity -  do several measures give the same result.
 

Oct 8
Measurement Chapter 3 in both books

Variables are characteristics or aspects that take different values among the things being studied.

In a questionnaire, often each question is a variable, but if it has a lot of choices, they may each be a variable

Are you Democrat, Republican, Independent or what?  One variable with three values.

Which of the following foods did you eat last week:
1. spaghetti   - values would be yes or no
2. soup
3 artichoke hearts
4.

Levels of Measurement.  What is our measurement really saying about the relationship between the values?

Dichotomous Measurement -   Two and only two categories.  "dummy variables" - we take a complex variable and divide it into a series of dichotomous variables.

Nominal Measurement.  Categories that could be put in any order.
      Catholic, Protestant, Jewish, Moslem, LDS, Buddhist, Episcopalian, Baptist
                       variable one, category of religion, variable two denomination.
            Illnesses:    adjustment disorder, borderline personality disorder, paranoid schizophrenic
               Crimes:   burglary, assault,
   Each individual should go into one and only one category on a variable, one value on a variable.
       Sorting people into categories must be reliable and accurate or valid.

Ordinal Measurement.   Here we have categories in a logical order.       Very short, short, medium, very tall, tall .  Often we take continuous variables and make them ordinal.    Income:   Under $20,000   $20 to 40,000  $40 to 60,000   $60000 plus.

Interval Measurement:   TEMPERATURE IN FAHRENHEIT OR CENTIGRADE, 0 degrees is not the absence of heat.

Ratio Measurement:    Income in dollars:  a continous numerical value PLUS a meaningful zero point.  Height in inches.
 
 

Intelligence or score on the test.   A number of questionnaire items can be combined to create a new variable.

Oct 1 -  General Review for first Midterm.  Some sample multiple choice questions.
 
 

FfREQUENCIES"  "OBSERVED FREQUENCIES"  "OBTAINED FREQUENCIES" or "DATA"	white	Black	TOTAL
Liberal	269	52	321
Conservative	381	58	439
TOTAL	650	110	760

eXPECTED MEANS WHAT i WOULD "EXPECT" IF THE NULL HYPOTHESIS WERE TRUE, IF THERE WERE NO DIFFERENCE BETWEEN WHITE AND BLACK RESPONDENTS IN POLITICAL IDEOLOGY.  Fe = RT * CT / GT
 

EXPECTED FREQUENCIES	white	Black	TOTAL
Liberal	321*650/760=274.5	46.5	321
Conservative	375.5	63.5	439
TOTAL	650	110	760

 

Percentages	white	Black	TOTAL
Liberal	row percent  269/321 * 100=83.8%  column percent  269/650*100=41.4%  total percent 269/760*100=35.4%		321
Conservative	col pct 381/650*100=58.6%		439
TOTAL	650	110	760

row poercent = cell frequency divided by the row total times 100
column percent is the cell frequency divided by the column total times 100
total percent is the cell frequency divided by the grand total times 100
83.8% of the liberal respondents are white
41.4% of the white respondents are liberal
35.4% of the respondents are liberal
 
 

                                           Race and Ideology Among Those Who are not "Moderate"

                                               White                Black

       Liberal                              41.4%                47.3%

       Conservative                    58.6%                 52.7%

      Total                                100%                  100%
           N =                             (650)                   (110)

             Chi square = 1.337   df =  1      p = .248, not significant

      Black respondents were slightly more likely to be liberal than white respodents, although the conservatives were the majority among both racial groups.  The difference between black and white respondents was not large enough to be statistically significant with this sample.  41% of the white respondents and 47% of the black respondents were "liberal".
 

Sept 29- for the first exam, you should know how to compute:  observed frequencies, row percent, column percent, total percent and expected frequencies for each cell in a cross-tabulation table.

25 men agreed
17 men disagreed
55 women agreed
35 women disagreed
 
 

Observed Frequencies or Obtained Frequencies	Men	Women	total
Agree	25	55	80
disagree	17	35	52
total	42	90	132

Take these observed frequencies and enter them into the WEB chisquare calculator.  Enter the answers on the answer sheet and in WEBCT.
 
 
 

Expected Frequencies	men	women	total
agree	80*42/132=25.5	80*90/132=54.5	80
disagree	52*42/132=16.5	52*90/132=35.5	52
total	42	90	132

Expected Frequencies - rt *ct /gt
 
 

Difference Statistic-	men	women	total
agree	25.5-25 = 0.5	0.5	80
disagree	-0.5	-0.5	52
total	42	90	132

 
 

Square (Fo*Fe)2/Fe	men	women	total
agree	(.5*.5)2/.5 = .125	0.5	80
disagree	-0.5	-0.5	52
total	42	90	132

chi square = .125 * 4 in this case, because all cells happen to be the same, and I add up all four.
 
 
 

Observed Frequencies	Men	Women	Total
Agree	50  or 55.5%	60  pr 50%	110
Disagree	40  or 44.4%	60  or 50%	100
Total	90	120	210

 

Expected Frequencies - take the null hypothesis that there is no difference between men and women on this issue.  How Many men would you expect to agree?	Men	Women	Total
Agree	52.4%  expected would be .5238 * 90 = 47.1	52.4% expected would be .5238 * 120=62.9	110 - 52.4% of the respondents agree, or the probability of any one respondent agreeing is .5238
Disagree	47.6% expected would be .47619 * 90= 42.9	47.6%  expected would be .47619 * 120 = 57.1	100 47.6% of the respondents diasagree, or the probability of any one respondent disagreeing is .47619
Total	90	120	210

to get a percent, you take the cell frequency and divide it by the total, then multiply by 100
to get the expected frequency, we take the proportion and multiply it by the total.

If 65% of my students are women and I have 80 students, how many women do I have?  .65*80 = 52
I have 52 women in my class and I have 80 students, what percent are women?   52/80 * 100 = 65%

Is the difference between the expected and the observed greater than we would expect by random chance?  Or, is it "statistically significant"  To answer that, we compute the chi square statistic.
 
 
 

Septemper 22 -   "per cent"  cent means 100.  Per cent is a ratio, with the denominator being 100.  A rate.  We have other rates, such as per 1000 or per 100000 or even per million or per billion.

The problem with percents is knowing the base, and how it  adds to 100.   What are the other components of the total.

men       55 agreed
women   33 agreee
men 27 disagreed
women 42 disagreed

Cross-tabulation Table.  Gender is the Independent VAriable or causal variable, so we usually put it in the column.  We put the Dependent Variable in the row.
 
 

Observed Frequencies	men	women	Total
Agree	55	33	88
Disagree	27	42	69
Total	82	75	157

Here is how it ought to be:
 

Column Percents	Men	Women	Total
Agree	67.1%	39.1%	52.2%
disagree	37.5%	60.9%	47.8%
Total	100%	100%	100%

There are three ways to do the percents.

In the row percent, the total is the number in the row which is used as the base.
In the column percent, the total is the number in the column which is the base.
In the total percent, the total is the grand total which is the base.

Here are the answers to the in class exercise which we did in class:
 

What are the two variables here?  Gennder, the other is opinion on figure skaters..
Values on gender are M and F, the values on opinion are agree, disagree.

Which is the IV?  GEnder, so we want it in the column.
 

observed frequencies	men	women	total
agree	85	95	180
disagree	15	25	40
total	100	120	220

Answer the following questions:

1. What percent of the men agreed?  85/100 = 85% "of the men agreed"
2. What percent of the women disagreed?  25/120*100= 20.8%
3. What percent of those who agreed were men?  take the number of men who agreed, 85, and divide it by the number of respondents who agreed, 180, times 100= 47.2%
4. What percent of those who disagreed were women?  25/40*100 = 62.5%  "of those who disagreed were women"
5. What percent of the respondents agreed?  180/220*100 = 81.8% "of the respondents agreed"
6. What percent of the respondents were women?  120/220*100=  54.5% "of the respondents were women"
7. Fill in the Table: (here we want column percents, based on gender, the IV)
 
 

	Men	Women	Total
Agree	85%	79.2%	81.8%
Disagree	15%	20.8%	18.2%
.	100% N =100)	100%  (N=120)	100%  (N=220)

 

September 17 -   "Junk Science"  How do we know?  Cold Fusion, Vitamin C, tobacco and cancer, does smoking "cause" cancer"  Go by the reputation of the scientist.  Where it is published.  Replication, somebody should be able to repeat your work.  Gun control.  John Lott = "more Guns Less Crime"    49% of the people believe the government should help with health care.
 
 

	Liberal	Moderate	Conservative	Total
Government Should Help	2	15	2	30
In the Middle	0	8	1	13
People Should Care for Themselves	0	0	0	0
Total	2	25	3

based on the cross-tabulation of ideology and opinion on health care funding, "61.8% of the liberals believe that the government should help with health care."   Number of Liberals = 474, Number of liberals who believe the government should help is 293.  Percent is 293/474 * 100  [only those who answered both questions are included, the "missing answer" people are dropped].

September 15 -
How do we know if our hypothesis was "correct" or "supported".  Most hypotheses involve a relationship between two or more variables.
Two steps:
1.  We want to know if the relationship is something other than random chance.  That is measured by the "propbability" figure that the computer gives you.  This tells you the probability that the relationship might have occurred by chance.  We want this to be less than .05.  That means we are 95% sure.  p<.05.  The asterisks on may Microcrase pages are a quick way of detecting this "*".
2.  Once we know we have something other than random chance, is the direction and strength of the relationship as we hypothesized.  To detect this we look at the percentage differences or the correlation coefficient or another measure of the strength of the relationship or the nature of the relationship.
 
 
 

Distinction between Experiments, Surveys, Field Research and Aggregate or Comparative Research, e.g. "time series analysis." Margaret Mead's classic work Coming of Age in Samoa which was extremely influential although Derek Freeman claims she was deceived by a hoax.  This has led to a long debate, with people generally believing who they choose to believe.
 

Purpose of Study	Preferred Design	Advantages/Disadvantages
Exploratory	Field Observation/Focus Groups/Case Studies	Gain new ideas and insights.  Samples are small and insights are subjective.
Description	Survey/Analysis of Trends in Existing Data Sets	Large samples gathered efficiently, makes use of existing data sets. Observations may be superficial.
Explanation	Experiment	Samples may be small, results limited to experimental conditions, may not reflect real life, not all variables can be experimented with
Explanation	Aggregate or Comparative Research	Treats problems not susceptible to experimentation/Good for looking at trends over time/Disadvantage:Statistical methods cannot prove causation.
Description	Content Analysis	Used to study documents or television or radio programs.  Can be used to study trends over tme.  Depends on coding for accuracy.

September 12 - Supplementary reading in Trochim on the structure of research.  You may prefer his "hourglass" metaphor to the circular one on page 14 of our textbook.  Discussion of Alumni Survey from 1995, to be repeated this year.  First Draft of Survey for this Semester.

September 8 - How does social science differ from other ways of thinking:  poetry, philosophy, theology, physical science? Three approaches to knowledge. Another approach to knowledge is aesthetic, creation of beauty.  Social science is generally pretty weak in this area, although occasionally the writing has literary value. W.H. Auden's poem Under Which Lyre is  an aesthetic attack on social science and other applied sciences...
Social science begins with concepts.  These are represented by words, but the important thing is the idea the word represents. Thus:  mother, madre, mere and mãe are all the same concept.  Numbers are also concepts, and they are perhaps the most fundamental.  The small integers are especially important, especially Zero and One (or nothing and something).  Social science is not the only thing to begin with concepts or words.  The Bible says In the beginning there was the Word, and the Word was with God, and the Word was God. In the Greek text, "word" is "logos" which is a thought element or concept.
We can define concepts any way we want, but there are good and bad concepts.  A good concept helps to organize reality in a useful way.  It leads to useful generalizations or theories.  Theories are general statements about relationships between concepts.  We can find a good list of sociological concepts by going to survey research archives, where concepts are translated into survey questions.  Check the General Social Survey and the Eagleton poll.Criminal justice concepts can be found on the  Bureau of Justice Statistics WEB site.
 I said there are also bad concepts.   For an example of one I think is bad, click on virtropy.  What's wrong with this concept?   Recently there has been much controversy over racial concepts or categories.  Is there such a thing as "race"?  What is it?  What races are there?   Census Document on Racial and Ethnic Categories.  Racial categories in Latin America.   Other concepts we can consider are: poverty, power, crime, murder, race, IQ, liberalism/conservatism, homelessness. Or we could look at Personality Types as defined by Carl Jung and Measured by Isabel Meyers-Briggs. There are also techniques such as concept mapping that can be used to develop concepts.

September 5 - Discussion of the IntroMicrocase online assignment.  This assignment is an abridgement of the Introductory Exercise on pages 13 to 18 in the Workbook.  You may wish to first do the exercise and jot the answers down on paper.  But to complete the assignment, you should do the "Microcase Intro" Online Quiz in WEBCT.  This quiz can be taken up to three times, and the highest score counts.  There are a few items not included in the paper version, so it is best to keep Microcase open in one window and Webct in another while answering it.  The Quiz can be taken any time up to September 15.