Grading Formulas:

Quizzes and Assignments=
([Enrolling]+[Microcase Intro]+[Workbook 1]+[Workbook 2a]+[Descriptive Statistics]+[Workbook 3]+[Sampling]+[Review for Midterm One]+[Workbook 5]+[Workbook 2b]+[PercentQuiz]+[Workbook 7]+[Workbook 8]+[Excel Regression]+[Review for Midterm Two]+[Human Subjects Letter]*2+[Crime Drop]+[Library Assignment]+[Historical Trends]+[Data Fair]+[Multiple Choice Review for Final]+[Statistics Review for Final])/23

Final Exam =  ([Final Multiple Choice Items]*0.75+[Final Statistics Items]*0.25)

Total Score =   ([Attendance]*0.1+[Quizzes and Assignments]*0.2+[Grade on Midterm One]*0.2+[Grade on Midterm Two]*0.2+[Final Exam]*0.3+[Extra Credit])

December 12 -  Review for the final.  The exam will be comprehensive, covering much of the same material that was on the first two examinations.  Reviewing these will be helpful.  The "Review Glossary" at the end of each chapter is very useful for getting definitions of terms - if you are still unclear about them, read the material in the chapter itself and in these notes. 

 In addition, there are two reqired quizzes intended specifically for reviewing for the final:  Statistics Review for final,  Multiple Choice Review for the Final.  The two review quizzes for the midterms  Review for Midterm Two and  Review for Midterm One will be open until the final, and you may find it useful to take them again.  There is also an Optional Levels of Measurement Quiz which is not required but useful for review.

Another tool is the Statistics Overivew that was prepared for the October 21 Statistics Review that was required for students who did poorly on the statistics items on the first exam.  I think our time today can most productively be used to go over this.  We will also go over the Statistics Review for the Final quiz. 

We also need to do course evaluations, and anyone who did not complete the Student Survey last week should do one today.

December 7 and 9 - extra credit presentation day.  There will be a few questions based on the presentations, check the Discussion Board for hints on these.

December 5 -- quiz make-up and extra credit preparation day

December 2 -  We attended the Camden Data Open House.  There is a make-up assignment for anyone who missed it.  This links to the WEB site that can be useful for anyone who wishes to refresh their memory of the presentations.

November 30:  Organizing and Giving Presentations.  Guideline for Oral Presentations.  One should be prepared, but usually not to the point of reading a text.  Two things are very useful:

1. Clear, easy to understand graphics
2. Bullet points summarizing the key points you want to make
   

The standard software for preparing presentations is Microsoft Powerpoint, and it has a lot of  useful features including the ability to incorporate photographs and video and to "animate" your bullet points (make them come up one at a time).  Here is an example from the 2005 Data Fair:  Camden Safer Cities Initiative.  At the same time, many people find powerpoint rigid and boring.  Here are some links on Powerpoint in case you want to pursue it.    Powerpoint logic.  Sontag on photographs.    Gettysburg Address in Powerpoint.  Breaking Up by Powerpoint. There's Something About South Jersey.  Trends in Camden:  www.invinciblecities.com.   Making Powerpoint Presentations.   Powerpoint Tips.  PowerPoint Tips and Tricks - includes a History of Powerpoint. 

One can put much the same material in Word, which is easier.  Just put each table or chart on the top of a page, with the bullet points below it.  Don't fill more than half a page so that it can all be viewed at once.  You won't be able to animate the bullet points. A sample presentation in Word is available.   If you wish to do Powerpoint, a sample powerpoint presentation is available that you can use as a template.   If you open it in Explorer, you will need to save it to disk and open it in PowerPoint to be able to edit it.  You can then  just replace the graphs and text with your own material, saving as much of the formatting as you wish to use.

You can creat graphics in Excel or Microcase and paste them into Word or Powerpoint.  If you do Microcase graphics, a lot of extra space will come out around them.  If you open the "picture" tool bar in Word, you can crop the graphics so they fit better on a slide or screen. 

Now, let's look at some of the ideas people have submitted.  I asked you to identify variables and a data set or data source.  We must limit ourselves to available data.  Some data sets are appropriate for some topics, others are not. 
 

November 28: 

Experimental Research.   Experimental Designs.  See the graphs in the book or on Trochim's WEB site:  Types of Designs. 

Essential characteristics:

1. Two or more groups are matched, usually by random assignment, sometimes by a kind of stratified random selection, e.g., an equal number of men and women or black sand whites in each group.  But the key is random assignment so that the groups can be assumed to be the same on all variables.  "Quasi-experiments" are when we use groups that are pretty much the same but we didn't assign people at random
2. The Independent Variable is "manipulated," i.e., it is applied to one group and not to the other
3. Change in the Dependent Variable is measured

Experiments can be done:

1. In laboratory settings with volunteers, e.g., student volunteers
2. In institutional settings such as prisons, hospitals, rehabilitation centers, etc., where people are assigned to treatment groups
1. New drugs and medical treatments generally must be shown to work in experiments before they are approved for use.  Often, treatment is compared to a placebo.  These experiments are usually "double-blind," to control for the psychological effects of knowing one is getting treatment.  This is a way of controlling subject bias and experimenter bias/
2. In criminal justice, one might do an experiment comparing a "half way house" to drug treatment program to a prison term for offenders.  To do this, you would have to get the judge to assign offenders to different programs at random.  Ethical issues are raised here and there are likely to be objections
3. Occasionally in natural settings, for example
1. welfare reform experiment, assign some recipients to the old program, some to the new.  This didn't work very well, there were errors in the group assignments and the women often forgot which group they were in anyway
2. vaccination experiments
3. guaranteed annual income experiments

Although logically experiments are the most rigorous way to test causal hypotheses, there are practical problems:

• It may be hard to manipulate the independent variable effectively, it may not have enough importance to people that they notice it
• Experimental conditions may not be realistic enough, e.g., the Milgram experiments having people apply electric shock to people, experiments that simulate being in prison.  An experiment is not the real world and people know it.  This is called external validity, does the experiment match real world conditions
• There may be problems of internal validity, difficulties in carrying out the experiment:
1. "History" effects - the world changes during the experiment, people get older, more mature, they are effected by things in the real world
2. Maturation, people get older, learn more
3. Testing effects, taking the pretest measure effects people, causes them to change.  Sometimes we have a matched but untested control group that is measured only after the experiment.
4. Instrument effects, the testing instrument may change.  You can't use the same exact test sometimes because people will remember it, so items change
5. Regression to the mean, just by chance the people who got extremely high or low scores on a pretest are likely to get more average scores on the second test.
6. Subject "mortality" - we may lose people.  This is especially a problem in testing things like drug rehabilitation, it works for the people who stick with it, the failures drop out
• Ethical concerns:  people may not be willing to be experimented on, or it may be harmful to subject them to experimental conditions, e.g.,
1. Tuskeegee syphillis experiment denied some men penicillin.  You can only deny an experimental drug if you are not "certain" that it works or if the condition is not serious, e.g., common cold research
• A big strength of experiments is resolving questions that involve different recollections of events, e.g., children's reports of abuse.  You don't know what "really" happened and people disagree on how well they accept the recollections of different people.  In an experiment, you know what really happened, so you can check the accuracy of perception.  We find that children often remember things that didn't really happen.   "20/20 report on Child Abuse experiments (VIDEO shown in class from an ABC News 20/20 show aired October 22, 1993, hosted by Hugh Downs.   Transcript available at www.transcriptstv.com) demonstrates false memory because we know what really happened since it happened in a controlled experimental setting.  This is much more difficult to establish in real life case histories:  Loftus:  Who Abused Jane Doe?  There is other information online on the Kelly Michaels case and other cases.

November 22:   open quiz day - all WEBCT quizzes will be available.  No regular class.  Happy Thanksgiving.

November 21:   News on Camden Crime.    Camden Crime Powerpoint by Camden Safer Cities Initiative.
 We will discuss creating presentations in Powerpoint and with other media.  Some examples we will view include an essay "September 11, 2001:  A Turning Point for America's Future" posted in in html format and  also available in powerpoint format.   Powerpoint logic.  Sontag on photograph.    Gettysburg Address in Powerpoint.  We also discussed materials from  http://www.futureswatch.org/, including the Trends Timeline chart available there in flash format.  Data for presentations can be obtained from Microcase files, or from statistical sources such as the Statistical Abstract of the United States.


November 18:     Trends in Camden:  www.invinciblecities.com.     Presented at a conference at Rutgers-Camden this morning.

We reviewed the Historical Trends assignment and viewed an example  Trends in Homicide Rates by my teaching assistant Fulano de Tal.  As an extra-credit project, this could be enlarged into a longer paper.
Here is an online paper treating the same topic: Alcohol Prohibition and Homicide.

A longer paper by the same author can be found through the library or through Google Scholar:

Miron, J. (1999).  “Violence and the U.S. Prohibitions of Drugs and Alcohol,” American Law and Economics Review 1-2, 78-114.

November 16.  We discussed the Library Research Assignment and the Required Reading Here material.   We viewed a  Sample Paper in APA Style. from the Penguin Writing Manual used in the writing intensive courses.  We also looked a the following papers, one of which is on the open internet, the other in a scholarly journal:   Web Paper on Alcohol Prohibition and Homicide.  Journal Article  Trends in Homicide Rates.

November 14 - Discussion of   The Crime Drop in America: Disaggregating Violence Trends.   We went over the WEBCT quiz on this article.

November 11 -   This was a discussion of the midterm exam.

November 9 - we went over the Human Subjects exam.  Some answers are available in our Discussion List.

November 7 -  Second Midterm Exam.  There were 38 multiple choice items (after two were eliminated) and 11 statistical items.  Grades were computed as follows:

Quizzes and Assignments = ([Enrolling]+[Microcase Intro]+[Workbook 1]+[Workbook 2a]+[Descriptive Statistics]+[Workbook 3]+[Sampling]+[Review for Midterm One]+[Workbook 5]+[Workbook 2b]+[PercentQuiz]+[Workbook 7]+[Workbook 8]+[Excel Regression]+[Review for Midterm Two])/15

Grade on Midterm Two = 0.75*[Midterm Two Multiple Choice]+0.25*[Midterm Two Statistics]

Predicted Grade = ([Attendance]*0.1+[Quizzes and Assignments]*0.2+[Grade on Midterm One]*0.35+[Grade on Midterm Two]*0.35)+3    [I added three points to raise the "curve" because I expect most of you to be doing better by the final.]

November 4  Review for the second midterm.  The exam is comprehensive in that it may include anything we have covered, but it will focus on the material after the first midterm including chapters 5, 6, 7, 8, 9, and 11.   We have not covered chapter 10 on Experimental Methods yet.  The Review Quiz for Midterm Two is designed to help you to review and is required.  Reviewing the other quizzes we have done since the midterm is also recommended, along with reviewing these class notes and the Review Glossaries at the ends of the chapters.


Looking at the graph of homicide trends, we can say that the  regression coefficient for the years from 1910 top 1930 was positive, whereas for the years from 1032 to 1942 it was negative.  For the entire period from 1910 to 1960 the regression coefficient would be close to zero because the points do not fit a straight line at all.  The R-squared would be very high for the period from 1910 to 1932 because there was a strongly linear trend.  For the period from 1910 to 1960, the R-squared would be cloze to zero because the trends do not fit the regression line at all.

The statistics questions will cover percentages, expected frequencies and regression.  The percentage questions will be very similar to the ones you did as an exercise.
Here are some sample regression questions that we will go over in class on friday. 

    Adult Drug Arrests in Thousands                                   
2000    1,375.6        SUMMARY OUTPUT                           
2001    1,384.4                                   
2002    1,352.6        Regression Statistics                           
2003    1,476.8        Multiple R    0.839978121                       
2004    1,551.5        R Square    0.705563244                       
                       Adjusted R Square    0.607417659                       
                       Standard Error    52.38970637                       
                       Observations    5                       

                            Intercept    -87500.66001
                            Coefficient  44.42000001
                                       
                     
Question One:       Assuming a linear trend, how many drug arrests would you predict for 2010?       To answer this, multiply 2010 by the regression coefficient, 44.42 and subtract the intercept, -87500.5501                                

Question Two:   How many would you predict for 2000?       Same as question one, except use 2000                                

Questioni Three:  How much does the number of adult drug arrests go up each year?     The answer to this one is just the regression coefficient 44.2, since it goes up that much when multiplied byh one.                                  

Question Four:  How many adult drug arrests were there in 2004? This requires knowning that we are talking in millions so the number is 1,551,500                                       

Question Five:  If we extend the trend back to the year 0 (when Jesus was born) how many arrest would there be?                   when the year is zero, you multiply it by the regression coefficient, getting zero.  So the answer is the intercelt, -87500.5  
Question Six:   What percentage of the variance in adult drug arrests can be "explained" by knowing the year?                
The answer to this is the R-square.
Question Seven:   Draw a time series graph showing the observed data and the regression line projected to 2010.                                   
This requires drawing by hand the same kind of graph you would produce with the scattergram procedure in Excel.

October 31

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.

Manifest Content - what's it's about on the surface
Latent Content - things that we infer about the content, e.g., does the writer sound angry?  Indignation, sexy?
A Content Analysis Study of Editorial Cartoons.
 A Content Analysis of Internet-Accessible Written Pornographic Depications.
A test of hand-eye coordination
http://www.coorslight.com/iceswipe

We can use the content analysis study as an illustration of many of the basic concepts from the first part of the semester that will be on the second midterm and again on the final.  We can use the definitions in the Review Glossaries in the textbook.  I am not going to repeat these  definitions in the notes.

October 28

Some examples of field resarch:
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.

Coming of Age in New Jersey. 

The Corner.     Memoirs:  Frey dispute with Oprah. 

Black American Students in an Affluent Suburb.  by John Ogbu. 

Commentary on Ogbu's research. 

Many scholars who have disputed those findings rely on a continuing survey of about 17,000 nationally representative students, which is conducted by the National Center for Education Statistics, an arm of the federal government. This self-reported survey shows that black students actually have more favorable attitudes than whites toward education, hard work and effort.

But that has by no means settled the debate. In the February issue of the American Sociological Review, for example, scholars who tackled the subject came to opposite conclusions. One article (by three scholars) said that the government data were not reliable because there was often a gap between what students say and what they do; another article by two others said they found that high-achieving black students were especially popular among their peers.

"It's difficult to determine what's going on," said Vincent J. Roscigno, a professor of sociology at Ohio State University who has studied racial differences in achievement. "'I'm sort of split on Ogbu. It's hard to compare a case analysis to a nationally representative statistical analysis. I do have a hunch that rural white poor kids are doing the same thing as poor black kids. I'm tentative about saying it's race-based."

Indeed, Professor Mickelson of the University of North Carolina found that working class whites as well as middle-class blacks were more apt to believe that doing well in school compromised their identity.

All these years later, Professor Fordham said, she fears that the acting-white idea has been distorted into blaming the victim. She said she wanted to advance the debate by looking at how race itself was a social fiction, rooted not just in skin color but also in behaviors and social status.

"Black kids don't get validation and are seen as trespassing when they exceed academic expectations," Professor Fordham said, echoing her initial research. "The kids turn on it, they sacrifice their spots in gifted and talented classes to belong to a group where they feel good."

Frey Dispute with Oprah        Dutch:  Fictionalized Reagan Bio.  NY Times review. 

October 26.  We will discuss the use of graphs to communicate statistical data.  The basic types of graphs to be discussed are:  pictographs, line plots, pie charts, map charts, histograms, bar graphs, line graphs, frequency polygons, scatter plots, stem and leaf diagrams and box plots.   These are described on the Mathland Website. You should know the basic descriptions of each kind of graph, their advantages and disadvantages and the kind of data required for them.   The stem and leaf and box and whiskers graphs are better explained on Steve Simon's page. Michael Friendly's Gallery of Data Visualation is a great collection of interesting graphs.  Many of the graphs are more complex than one can do in Excel, but lots of good graphics software is available, such as OriginLab..  The Stem and Leaf you can just type, you would have to draw a boxplot or find software that does it.   Variatons of line graphs.  Drug War graphs. 

October 24 -  We discussed the use of Excel to make time series graphs and linear projections.  There is a full explanation and links to resources in the assignment file. 

October 21 - This is the Statistics Review class required for students who received less than 80% on the statistics questions on the first midterm.  The Statistics Overview, Review assignment and review assignment Answered, are available. 

October 19 Comparative Research Using Aggregate Units, Chapter 8 in the text.  This research method uses data about social or geographic units.  Consistent criminal justice statistics are important for evaluating CJ policies.  Thorsten Sellin, a professor at Penn, was instrumental in getting consistent CJ statistics established. 

Comparative methods are particularly useful for studying change because we can get data about trends over time.   This is available in the computer center on the networked Windows computers (click on Statistics and Microcase on the Windows menu, then open "Microcase Curriculum Plan 2003-2004 and load the TrendSmp data set.  Our next assignment requires using this data set in the computer lab..

Some concepts:

Rate:  A statistic that reduces numbers to a common base.  The base is often, but not necessarily, the total population in an area.  If we are looking at voting participation, we might compute rates using the base of the number of adults 18 or over.  If we are trying to predict an election, we might use a base of registered voters. 

A crude birth rate is the number of births per 1,000 population.  Fertility rate is the number of births per female during her lifetime. 

Time Seriesor Trend analysis:  uses time periods as the unit of analysis, looks at how things change over time often in one case.   A lagged time series takes into account the time it takes for one variable to influence another, thus incarcerations in one year might be related to crimes in the next year.  We can find examples on the Bureau of Justice Statistics WEB site. 
We can also do trend analyses with the  Historical Trends module in Professional Microcase which is available in the labs on campus, or which you can install at home.  Here are some Trend Graphs taken from the "Historical Trends" module in the Professional Microcase.   We will also be learning to make trend graphs with Excel. 

Cross-sectional analysis compares a number of cases at one point in time.  These are often states, but they could be counties or precincts or countries.  We have already done quite a bit of this with the mapping and scatterplot programs in Microcase.  This can also be done with Excel, where cross-tabs are called "pivot tables".  The paper we examined earlier on Capital Punishment and Homicide has both trend and corss-sectional graphs.

Reliability:  are statistics computed the same way in different geographic units or different time periods.  This causes all sorts of problems - it is better to imporve statistics, but doing so causes us to lose comparability. 

Validity:  do the statistics measure what we want them to measure.  Crimes reported to the policy are not a valid measure of the amount of actual crime, especially for crimes that are often not reported. 

Case oriented vs. variable oriented.  The case oriented approach is more qualitative, although quantitative trend data can be used.  The variable oriented approach assumes that the same variables are causally related in the same way in a large number of cases, e.g., "capital punishment" and "homicide rates" in a number of states or countries. 

Outliers:  especially in variable-oriented research, it is important to look for exceptional cases that are very different from the norm.  These tend to cause a disproportionate impact on our results. 

Lagged:  Using statistics from past years to predict events in current years.  This is done because our theory says that causal linkages take some time to take place.

October 17:  Survey Research, chapter 7. 

Surveys are used to measure proportional facts - how many people believe or do X in a given population?  Sampling is one issue, and we've discussed that already.  Once you have a sample, you have to decide what questions to ask.  Key to survey research is that everybody gets asked the same questions, at least if appropriate.  The kind of questions used depends on what you need to know:  Open ended vs. closed ended.  multiple choice vs. short answer or essay.  Questions have to be clear and unambiguous.  Reliability and validity depends on the respondents' willingness and ability to answer - people can't answer questions if they do not have an opinion or do not understand the topic under discussion.

The next step is interviewing.  This can be done by telephone or in person or by mail or on the Internet.  Sometimes computers do the interviewing by telephone.
Guidelines for Interviewing

1. Study the questions carefully to make sure you understand them and can read them easily and conversationally.  Practice by interviewing yourself or your friends or relatives.
2. Be enthusiastic when the respondent answers the telephone.  The hardest part of an interview is the first ten seconds.  Make it sound like something fun.  You are free to reword the introduction in a way that is natural to you.
3. Don't ask permission, just get started with the first question.  Of course, people have a right not to participate.  If they are just busy, however, ask when you can call them back.
4. Be enthusiastic and appreciative - you truly appreciate the respondent giving his or her opinion.
5. Never be critical or sarcastic.
6. Be neutral, don't give your own opinion or even hint at it.  You are interested in their opinion, not in sharing your own.
7. This is tricky:  read the questions as they are written, but make it sound as if you are speaking informally.
8. If the respondent is not clear about a questions and asks you to explain it, the best thing is to repeat it slowly.  Usually that is all they need.
9. Give the respondent time to think.  Often they will give an answer after a few moments' reflection.
10. If the respondent doesn't have an answer to an item, just go on to the next item.  Don't try to extract an opinion that isn't there.
11. Be sure to thank the respondent for participating.

Once the interviews are done, the data has to be inputted into a computer system and cleaned.  This is discussed in the introduction to Exercise 7 in the Workbook.  I'll use that as an example in class, using the "Test" data file.  We will not learn to do these tasks in this course, because it is specific to the particular software you are using.  It is a matter of reading the instructions carefully.

Finally, you get into analyzing the data, which is what we have been doing already with data collected and cleaned and recoded by others for our use.  Still, we need to make a judgment about how good the data are, so part of our analysis is judging reliability and validity.  Also, we recode variables for our own purposes and construct scales or indexes or other composite measures. 

  October 14:  Research Design.  How research is organized or structured to accomplish different ends.  This depends on the purpose of the study as well as on practical matters such as resources and researcher preferences.  Your book discusses four "basic" types of designes in chapter 6, which is a good beginning:

1. The experiment  -  apply the IV to one group, compare the outcome to a control group
   
2. Survey Research -  ask a list of standardizes questions to a representative sample of people
   
3. Field Research - go out into the world and observe what actually goes on
   
4. Aggregate or Comparative Research - analyze statistics collected by government or other organizations
   

There are many variations on these basic forms, some of which are discussed in other chapters (e.g., the chapter on Comparative Research discussed case study research briefly).  Also, many researchers combine several different research designs in the same study to get different kinds of information.  In the table below I outline the advantages and disadvantes of several research designs for different purposes.

Purpose of Study	Design Alternatives	Advantages/Disadvantages
Exploration - To get some new ideas, or at least ideas that are new to you.	1. Literature Review - library research, almost everybody begins here. 2 .Field Observation - Go into the natural setting and observe what is going on.  You may talk to people and ask questions as well, but the really unique aspect is observation.  3.  Laboratory observation:  recruit people into a laboratory situation and observe how they behave when asked to do different tasks.  May be done with children or adults, often with a one-way mirror. 4. Focus Groups - Group interviews lasting about an hour and a half.	1.  Get insights of others.  Avoid reinventing the wheel./ Tends to repeat the past, not generate new ideas. 2.  Get new insights in natural setting/ Difficult and time consuming, small sample.  Access difficult. 3.  Good for studying interpersonal patterns so long as the artificial setting doesn't change them. 4.  Good for generating insights and new ideas;  sometimes one person dominates the group.  Difficult to generalize to large population.
Description - To get accurate and relatively precise information, especially about large groups or	1. Summary of Trends in statistical data available from government or other sources.  Data banks of surveys are available, many other kinds of data also.  We can describe patterns in the past or in other places, look at trends 2. Conduct your own Survey - Questionnaires or interviews.  Often on the telephone. 3.  Content analysis - Looking at media as a source of data:  tv shows, letters to the editor, newspaper articles.  Written documents.  You can go back in time. 4. Case Studies - based on documents, interviews or sometimes observations.  Often several cases are compared, a method we might call small-n comparison.	1.  Excellent data, especially for trends over time/  Limited to questions asked by others. 2.  Ask your own questions, choose your own sample.  Reliable, replicable  results.   Limited to topics people can answer accurately.  A lot of work or costs money to get professionals to do it. 3.  Unobtrustive, allows study of media./ Limited to topics that involve published media. 4.  Provides holistic, complex understanding but it is difficult to generalize.  Most interesting when several cases are compared.  The cases might be corporations or countries or police departments.
Explanation.  To answer questions about cause and effect.	1. Experiment - In an experiment we manipulate the independent variable.  The independent variable is the "cause" .  Then we measure the dependent variable or "effect" both before and after on experimental and control group. 2.  Comparative trend analysis, see how the trends change under different conditions.  This is often done with data collected by others.. 3.  Multivariate crosstabulation analysis analysis using cross-sectional survey data.  4.  Multiple regression analysis of data which may be from psychological tests,  or economic indicators or government or medical data sets (including criminal justice data).  This is (sometimes called econometrics or path analysis or structural equation modeling).  5.   Computer simulation;  building computer programs that simulate real life with actors or forces that interact with each other.  Used for difficult topics such as predicting the weather.	1.  Best method of proving causal relationships./ Hard to maintain rigor of design (internal validity) and to generalize beyond the limits of the experiment (external validity).  Serious ethical and practical limitations. 2.  One can observe which trends go together and draw causal inferences based partly on the data and partly on other knowledge. 3.  Causal relationships can be tested by using statistical controls to control for test variables.  Results depend on how good the sample is of various sub-groups, how good the questions are, and other things.. 4.  Data sets must include good measures of all relevant variables and wide range of data.  Not valid unless the models can be shown to predict trends in fresh data.  Results have often been contradictory because there are so many ways to manipulate the data. 5.  Difficult to get the models to work or to know if they work for the same reasons as real world processes.  May be the future with more powerful computers and software.

We may find it useful to look at some examples.  Note that many studies incorporate a number of methods.  For example, a study of traffic law enforcement.  Or Felton Earls's work that we examined a bit earlier.  Research on   the measurement of romantic love.

October 12:

 Let's go over the computation of row, column and total percents and also expected frequencies in cross-tabulation tables. .  For this purpose we will use a simple 2 by 2 distribution as follows.  The variables are gender and opinion on an issue, each of which has two values:

25 men agreed
17 men disagreed
65 women agreed
30 women disagreed
 
  The first thing we do is put them in a two dimensional table, as follows and compute the row totals, the column totals and the grand totals. 

Observed Frequencies or Obtained Frequencies	Men	Women	total
Agree	25	65	90
disagree	17	30	47
total	42	95	137

To get the column percents, we divide the cell frequencies by the column total, then multiply by 100 to get a per cent.  Thus, if I ask, "what percent of the men agree" the answer is 25/42 *100  =  59.5%.  The base of this percent is the number of men.  This is a column percent because the men are in a column.
If I ask,  "What percent of those who agree are men," the answer is 25/90 * 100 =   27.8%,.  The base of this percent is the number of people who agree.  This is a row percent because the people who agree are all in a row.
If I ask, "What percent of the respondents are men who agree," the answer is 25/137*100 =  18.2%.  The base of this percent is the total number of respondents.  This is called a total percent because the base is the total number of people.

 We can compute expected frequencies, based on the null hypothesis that men and women do not differ in their opinions.  We can compute these knowing only the marginal or total frequencies.  The easy way to compute them is to multiple the row total for each cell by the column total for that cell, then divide by the grand total.    Expected Frequencies - rt *ct /gt

You can see examples of these with the Percents, Expected Frequencies and Chi-Square Calculator (an Excel spreadsheet). 

This also calculates the chisquare statistic which is given by the formula  (ObservedFrequency-Expected Frequency)²/ExpectedFrequency.  You can then look this up in a table in the back of a statistic book to find out if the difference between expected and observed is "statistically significant".  

Expected Frequencies	men	women	total
agree	90*42/137=27.59	90*95/137=62.41	90
disagree	47*42/137=14.41	47*95/137=32.59	47
total	42	95	137

The following is an example I typed in class.  The material is red is new.  The items are the same as on the "percent quiz" assignment, but with different numbers. 
Consider the following answers to the question "I believe that marinated artichoke hearts should be the national vegetable."

65 men agreed                             

	Male	Female	Total
Agree	65	25	90
Disagree	85	105	190
	150	130	280

25 women agreed
85 men disagreed
105 women disagreed

Answer the following questions:

What percent of the men agreed?   PCT1 .           .   This is a colum percent because the men are a column.  65/150 * 100  43.3%
What percent of the women disagreed?  PCT2 .           . Also a column percent   105/130 * 100  = 80.8%
What percent of those who agreed were men? PCT3 .           .   This is a row percent.  The row is the Agree row, the total is 90.  the men who agree are 65   65/90 * 100  72.2%
What percent of those who disagreed were women? PCT4 .            .   105/190
What percent of the respondents agreed? PCT5 .            .    The number who agreed divided by the grand total.  90/280   32.1%
What percent of the respondents were women? PCT6 .         .

Fill in the Table:

Gender and Belief that the Marinated Artichoke Hearts Should be the National Vegetable

	Men	Women	Total
Agree	.	.	.
Disagree	.	PCT7 .      .	.
.	100%	100%	100%

This table asks for column percens because they add to 100%.  to get what % of the women disagreed, as asked for, divide the women who disagreed by the total number of women. 

	Men	Women	Total
Agree	PCT8.   .21.1	.   23.9	45
Disagree	.53.9	PCT9 .  .61.1	115
.	75	85	160

This is establishing a "null hypothesis" that gender and opinion do not matter.  The expected frequency is what we would "expect" on the null hypothesis that there is no relationship between the variables. The easy way to compute them is to multiple the row total for each cell by the column total for that cell, then divide by the grand total.    Expected Frequencies - rt *ct /gt

for the men who agree, the expected frequency would be 75 * 45 /160 =  21.1  THIS IS NOT A PERCENT.
MEN WHO DISAGREE 75*115 /160 = 53.9
   women who agree   85 * 45 /160  23.9
women who disagree 85 * 115 /160  =  61.1 


October 10

Looking at the example on page 139 in the workbook
In the sentence, "the higher their incomes the more likely people are to support freedom of speech," there are two variables:  income and support for freedom of speech.   Income is the "independent variable" and "support for freedom of speech" is dependent.  How do we graph that?  We put the DV on the right and the IV on the left and draw an arrow from the IV to the DV. 

We can test this bivariate hypothesis by using the regressioni procedure in Microcase.  We use COMSPK as the dependent variable and R.INCOME! as independent.  We find that the BETA is .162 and it is statistically significant. 

The statistic we want is a STANDARDIZED regression coefficient or "beta" .  These vary from -1 to 0 to +1 like correlation coefficients.  For a bivariate case, they are the same as the correlation coefficient.  We have established that the two are correlated (one of the three criteria of causation) and we are willing to assume that the income level came before the attitudes (our second criterion).  Now we have to test for "spuriousness" or whether it can be "explained" by some antecedent variable.    They suggest using education as a test variable: 
"this is really a spurious relationship because both variables are the result of education."   What kind of a variable is education in this causal model?  According to the statement, education is an antedent variable.  So if the correlation between income and "support for freedom of speech" disappears when we control for education, we would say that it has been "explained" and that it is "spurious".

See the example in the Excel diagram:



To use the regression method you must have interval or dichotomous variables.  if we hav nominal variables, we use cross-tabulation.  This gives us percentage differences.  An advantage is that we see the actual patterns, we don't assume that they are "linear". 

We usually put the Independent variable in the column and the Dependent variable in the row, although that is not statistically  necessary.  
Suppose we ask the question "what percent of the liberals believe the government should do more?   The base of the percent is given by the phrase "of the liberals" so it is the total number of liberals which is  463.  The numerator is the number of liberals who believe the government should do more, which is  145.  The percent we want is 145/463  *100  which equals 31.3%. 

October 3 - First midterm. Grades are in WEBCT.  The columns in "Grades" are the following:

Predicted Grade -  my weighted prediction of how you are likely to do at the end of the semester
Grade on Midterm One -  your grade on the midterm including both multiple choice and statistics items
Attendance -  your attendance grade
Midterm One Multiple  Choice - your percent correct on the multiple choice items (should correspond to the answer sheet I will give out in class)
Midterm One Stats -  your percent correct on the 8 statistics items on the midterm
Quizzes and Assignments - your average score on the assignments and "quizzes"
  -  others:  your score on each assignment -

 Grading formulas as follows:

The following formulas were used in computing the grades:

Quizzes and Assignments  =  ([Enrolling]+[Microcase Intro]+[Workbook 1]+[Workbook 2a]+[Descriptive Statistics]+[Workbook 3]+[Sampling]+[Review for Midterm One])/8

Attendance =  12 classes/.11  (one bonus class)

Grade on Midterm One =  [Midterm One Stats]*0.2+[Midterm One Multiple Choice]*0.8

Predicted Grade =  ([Attendance]*0.1+[Quizzes and Assignments]*0.2+[Grade on Midterm One]*0.7)

We will go over the test on Wednesday.  If you have fallen seriously behind at this point, you need to either make a real effort to recover or decide to withdraw and take the course again next semester.  There is still room to improve:  the second midterm is 20% and the final 30%.  Plus we will have a Quiz Make-up Hour on November 22.  There is also extra-credit, but this will only raise your grade 5 percentage points at most.  Extra credit is for people who want to do an individual project and present it to the class .

October 5 and 7
Graphs drawn on the board on October 7:  



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.

Example:  research on capital punishment.   Powerpont.      Paper on Capital Punishment and Homicide. 

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. 

Sept 30 - class was conducted as a chat room.  The transcript is available online. 

Sept 28     We discussed levels of measurement and some people seemed to still be confused.  I'll attempt a better explantion here.

The first and most important question is:   is the measure continuous or categorical?   This is important because continuous variables are required for the use of statistics such as the mean, standard deviation, correlation and regression.  With continuous measurement we have precise distances between the items measured, with categorical we just have them sorted into discrete categories.

If a variable is continuous, we can ask whether it is "interval" or "ratio".    Both of these have precise distance measurement between points.  In addition, ratio measures have a logically meaningful zero point.  With ratio measures, we can talk about ratios between variables, e.g., say that $50 is twice as much money as $25.   With interval variables, such as fahrenheit temperatures, we cannot make such statement.

If a variable is categorical, we can ask whether it is "dichotomous,"  "nominal" or "ordinal"

Dichotomous variables have only two categories.  These can be two natural categories such as "male' and "female"  or they can be artificial "dummy" variables, such as:   are you a Catholic or not;.  With dichotomies you can u se regression and correlation.

Nominal variables have more than two categories, but not in any order or with a measured distance between them.

Ordinal variables have the categories in a logical order  (from "lower" to "higher").  An

In answering questions about measurement, give the highest or best level of measurement that is justified.  Any variable that meets the criteria for a ratio variable also meets the criteria for an interval variable, but the criteria for a ratio variable are more stringent so we would say that it is ratio measurement.  Any ordinal variable also meets the criteria for a nominal variable, but if it meets the criteria for ordinal we say it is ordinal.

 Today will be a general review for the Midterm on Monday.  On Friday, we will have the online chat review (see the course home page).  If you have specific questions, this will be an opportunity to ask them.  After Friday's class, please use the Discussion List in WEBCT for any further questions.  The advantage of this over email is that everyone can see the answers.  I will log in over the weekend and answer questions posted there.

If you have not done so already, this is the time to read the textbook.  We have covered chapter One, Two, Three and Four.  There are also summaries of this material in the Workbook on pages 19, 67 and 91.  You should be able to answer the questions listed under "before you begin".  The test will be mostly multiple choice, so probably the most useful for reviewing is the "Review Glossary" at the end of each chapter in the text.  There will also be a page of statistics questions similar to those we have had in exercises: 
    *  computing the mean, median, standard deviation as explained in the handout on descriptive statistics. 
    *   computing margins of error and sample sizes as explained in these notes under September 23. 
     *  using regression equations to predict one variable with another, e.g., if you know someone's height you should be able to use a regression equation to predict their weight.  This is explained in the notes under September 12. 

I am not going to summarize the four chapters in today's notes, you can find the class notes for each day in this file.  You may find it useful to print out these notes.  You also need a copy of the descriptive statistics handout.  You may also want to link to some of the examples used in class, in which case you need to view these notes online.

There has been some confusion about the levels of measurement.  These are explained on pages 34 and 35 of the book.  It is important to understand that many variables can be measured at different levels.  Thus I could take height and put it into categories such as short, medium, tall in which case I would be using ordinal measurement because they are in order.  I could also measure it in inches or centimeters, which would be ratio measurement.  It is also important to understand that each of the statistics is appropriate for variables measured in some ways but not others.  Doing percentages and cross-tabulations makes sense for nominal or ordinal data. Chisquare is for nominal or ordinal data. Doing correlation or regression or means and standard deviations requires interval or ratio data.  We can make a broad distinction between categorical (nominal or ordinal) or continuous (ratio or interval) data.  The dichotomy is a special case because we can use correlation and regression with dichotomies, but we can also do percentages, cross tabulations and chisquares.
   


Sept 26
Scaling or index construction is when we use a number of items, such as questionnaire items, to measure a more general concept.  We can do this by adding them up (in which case your text would call it an "index", although many people still use the term scale) , or they may be ordered from lowest to highest (in which case it is a true scale as the term is used in your book).  Your test is an example.  I just add up the points, to measure the general variable "knowledge of research methods as covered in the first part of the course."  Another approach would be to rank the items from easy to hard and see which you could do.  This is tricky, because some people can do the hard ones and not the easy ones.  When we make an index or scale, we get measures that can be treated as interval, even if they are not strictly interval.  Scaling methods can be more precise, but these are not used as often in sociology or CJ because they are more difficult and the added information is not always needed.

Scaling methods include Thurstone and Guttman Scaling.  Likert or summative scaling is actually a method of "index" construction as defined in our book.  A powerpoint on Thurstone scaling. 
  For example, we could scale the seriousness of crimes.  There are various methods of measuring this. - paired comparisons means asking a sample of people to rate crimes based on their perceived seriousness.

A very popular test is the Myers-Briggs Type Indicator, based on Jungian personality theory.  You can takeseveral free versions of this and related tests online (the Wikipedia article).  .   

Many measurements of crime trends are based on scales that add together a number of crimes, e.g. "violent crime". , 2005. :

U.S. crime rate remains at lowest levels in years
Based on victim surveys, the incidence of violent crime is statistically unchanged from last year.
By Mark Sherman - Philadelphia Inquirer Sept 26, 2005


Sept 23 -
SAMPLING is used when we are interested in studying a population that is too large for us to study each individual.  The first step is to define the population we wish to make statements about, e.g. adults in New Jersey, probable voters, people convicted of felonies, graduates of our department.  We might want to study the entire population of the USA.  If we try to collect data from everyone, this is a census.  The Census Bureau does this once every decade, and misses a lot of people.  Everyone else does sampling, we select a cross-section to represent the population.  If you try to study the whole population, you often fail to do a good job.   Gallup:  How Polls are Conducted.

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%

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.

This formula is for  proportions or percents (if you move the decimal over two)
  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 our sample is stratified, this means we really have several sub-samples and we need the same size sample for each of them, regardless of the size.  For example, if we want sample white, black and Hispanic respondents and make statements about each group, we need the same size sample of both regardless of their size in the population.  Thus, if we need a margin of error of 5% for each of the three groups, then the answer is  3 * ( n = 1/ m² ).

If you need a margin of error for a mean score (an average such as income in dollars or scores on a test), you need to know the standard deviation (sd) and the sample size (N). Ignore any other information you are given, including the size of the population.
Use the following formula: M = 2 * sd / SQRT(N)

Suppose I sample 457 Camden residents and the mean income is $27,541  and the standard deviation is $3452

M = (2 * 3452 )/sqrt(457).  This result will be in dollars, not percentages. 

M =      6904        /21.378  =  $322.95.  

Confidence Interval:   I am 95% sure that the population figure is between:  $27,218.05 and $27,863.95

Terms:

Margin of Error:  How much a sample statistic is likely to vary from the population parameter.  We say that we are 95% sure that the sample is not off by more than the margin of error.  How this is presented in NY Times.  "19 out of 20" is another way of saying 95%. 

 Confidence level:  we always use a 95% confidence level.

Confidence interval:  the range within which we think a statistic would fall, e.g., if the margin of error is 3% and the sample statistic is 67%, the confidence interval is from 64% to 70%.  We are 95% sure that the true figure is within this limit.

All of this assumes a simple random sample, which means that each person (or other sampling unit) in the population has the same chance of appearing in the sample.  In practice, however, we often do not use simple random samples, for several reasons:

1. we may not have a list of the population.  If we do not, we first divide the sample into sub-groups of some kind (census tracts, blocks, classrooms, organizations, depending on the nature of the study).  We then sample the subgroups and list the populations in them .  This is called cluster sampling
2. We may be interested in differences between sub-groups of the sample and need to make sure we have enough of them.  In this case we select random samples of each of the relevant sub-groups, and weight the results appropriately.   This is called stratified sampling. 
   
3. Sometimes we just go down a list, which is called systematic sampling.  This gives the same results as simple random sampling, unless there is some systematic ordering to the list that causes a distortion
4. Sometimes we use non-random or "quota" sampling.  This is done for convenience, or because we just want to know what the range of differences is without putting numbers on them.

An example:  NY Times Poll on George Bush, Sept 2005.


Sept 21 -
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.
          Convergent validity -  do several measures give the same result.              
          Construct validity - does the measure perform as our theory says it should.  We use this when we have no criterion.   This is the most difficult, it is used when things are inherently difficult to measure. 

                  An example:  a study of UFO Abduction Status.

Sept 19 -  Measurement means putting observations into categories.  Often these categories are given numbers, although not always.  Sometimes we do this just to keep track of things, e.g., each American has a social security number, we have a library number, a student number, etc..  But often the numbers give us more information than that, e.g., the NJ driver's license gives height in feet and inches.  It also gives sex and eye color, which are described in words but could be given arbitrary numbers.  But the numbers given for height are not arbitrary. In some sciences, e.g., astronomy, numerical measurement has led to important insights, e.g, to understanding the motion of the planets.  This is because our observations can be summarized with mathematical equations that enable us to predict events.

 When we measure something, we need to be clear exactly what the measure means.  Especially when we use a number, we want to know what it means.  What is a number?  It is not so obvious as one might think.  Bertrand Russell said "A number is the class of all classes similar to a given class."  I.e., all sets of three have something in common, which we could call "threeness."
 
Levels of Measurement.  What is our measurement really saying about the relationship between the values?

Dichotomous Measurement -   Two and only two categories.  Can be a natural dichotomy or a  "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.
            Mental illnesses (DSMIV) e.g.,  adjustment disorder, borderline personality disorder, paranoid schizophrenic
               Crimes:   burglary, assault, murder.  What do these terms mean?  Look at the US Criminal Code.

  Each individual should go into one and only one category on a variable, one value on a variable.   For example:  What is your favorite food, we have a long list, but each person is allowed only one.
       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.  How about the day that the "temperature doubled" in New York City?

Ratio Measurement:    Income in dollars:  a continous numerical value PLUS a meaningful zero point.  Height in inches.
 
Scaling is when we use a number of measures, such as test scores or questionnaire items, to measure a more general concept.  This often allows us to move to a higher level of measurement.  For example, we can add up test score items them up (in which case your text would call it an "index", although many people still use the form scale) , or they may be ordered from lowest to highest (in which case it is a true scale as the term is used in your book).  Your test is an example.  I just add up the points, to measure the general variable "knowledge of research methods as covered in the first part of the course."  Another approach would be to rank the items from easy to hard and see which you could do.  This is tricky, because some people can do the hard ones and not the easy ones.  When we make an index or scale, we get measures that can be treated as interval, even if they are not strictly interval.  Scaling methods can be more precise, but these are not used much in sociology or CJ.  For example, we could scale the seriousness of crimes.  There are various methods of measuring this. - paired comparisons means asking a sample of people to rate crimes based on their perceived seriousness.

One of the reasons we have to be clear about levels of measurement is that the statisitcs we use depend on how the data are measured.

Statistics for Nominal Data:    Percentages and Chi Square   The percentages are descriptive (they summarize our data), the chi square is inferential (it tells us if we can generalize from our sample).   Survey data usually produces nominal (or ordinal) statistics.   Cramer's V is a correlation coefficient for nominal data, scores on it vary from 0 to 1, but there are no negatives since the data are not ordered.

Statistics for Ordinal Data:  The median is the only statistic we have covered that is specifically designed for ordinal data - it finds the case in the middle once all the cases are sorted in order.  There are correlation coefficients for ordinal data which you can find on the "statistics" page for crosstabulations (gamma, tau) but it is more common to use interval statistics (Pearson's r) or nominal ones (Cramer's V) with ordinal data.

Statistics for Interval Data:   Scattergrams, means, standard deviations, correlation coefficients.  Tests of statistical significance for correlations

Sept 16 -
Discussion of designing research projects.  How do we decide what to study?  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.

1. Selecting a topic.  Typical motives include:
1. Finding out something we don't know.  This may include something local, e.g., what do people in Camden think about the new Governor's actions, something that has been unresolved in earlier research, something that hasn't been studied because it is new, etc.  This is what the authors of your book mean when they say "research always starts with wondering."
2. Another purpose that motivates research is proving to other people that what we "know" is true really is true.  This is "advocacy" research, and it can be very one-sided and lead to sloppy work.  Often this involves causal arguments, proving "why" something happens.  This kind of research may not start with "wondering" but with "arguing."
3. Answering a question posed to us by our employer or by a client, applied research.  Here someone else really chooses the topic.
2. Formulating a Research Question.  This means formulating a "statement" which will involve variables.  We have an argument or story in mind at this point.
3. Defining the Concepts.  Usually not a lot of time goes into this stage of empirical research, but some people do write articles focusing on this, e.g., what does "race" or "poverty" mean, what is the difference between "sex" and "gender"  An example:  the measurement of romantic love.
4. Operationalizing the Concepts.  A lot of effort goes into this.  Quantitative  research means you have to measure your variables and a lot depends on having good measurement.  Sometimes this is difficult, e.g., measuring "intelligence" or "liberalism-conservatism" or "mental illness" or "crime rates (various kinds)".  Often we use standard measures created by the government agencies that collect statistics.
5. Formulating Hypotheses.  This is usually pretty easy.  There is a distinction between "null hypotheses" and regular hypotheses, which is explained on page 13.  It means testing the hypothesis that your hypothesis is not true.  Thus, you hope to "reject the null hypothesis" rather than "accept the (regular, not-null) hypothesis".  So far as I know, there is no word for the opposite of Null, it might be Substantive?  Type One Error:  accepting that a relationship exists when it doesn't.  Type two:  rejecting a relationship when it really does exist.
6. Making observations.  This is a major step unless we just get the observations from someone who already did the work.
7. Analyzing the Data.  This is "number crunching"  running data through the computer.  Of course, one can also analyze qualitative data from interviews or observations, but today even that tends to get quantified (content analysis).
8. Assessing the results.  This is really part of the analysis.  If the hypothesis doesn't work out, often researchers go back and change the hypotheses and pretend they knew all along what was going to happen
10. Publishing the findings. This assumes that you are doing "scientific" or "pure" research, much applied research is actually distributed only within the organization that paid for it.  This may be done in person, with a "power point" presentation.  Refereed publications:  you paper is sent to other specialists for review to decide if it should be published.  "Refereed journal."  Press release.   Publication can be online as well as on paper.  You publish the research so you can get credit, see your name in print, get promoted, and also so that you can inform others, and perhaps most important, so that other people can criticize or attempt to replicate it.    Usually people replicate research in the hope of overthrowing it, if you just find the same thing as before, there is less interest.  This cancels out a lot of the bias in social research, since there is usually someone with the opposite bias to correct it.

    Here are some samples we can look at:NY Times Poll on George Bush, Sept 2005.   Papers presented at the 2000 ASA meetings in Washington, a  Study of Tire-Crash Patterns (Word Format with Excel File Used to Reproduce Graphs.)  The controversy over a study on the effects of sex abuse. Compstat in the  NYC and Philadelphia  Police Departments.     The origin and development of the project on South Jersey's Identity that we workied on in this class in 2000.  Results are on my home page.  Last semester we worked on a survey of graduates of this department.  The Questionnaire is available online.  We did an earlier survey in 1995, a Report is available.  Contacts between Police and the Public.    The 2002 Final Report on the National Drug Control Strategy.    And the 2003 version - the emphasis on the goals has been lessened, with the excuse of discontinuities in data collection.  2003 Tables in HTML presentation form  . 

Sept 14  -  we went over calculation of descriptive statistics-  the Descriptive Statistics page was handed out in class - this shows how to do the calculations we need.

Sept 12 -   Regression Equations. 
We will discuss the research process using aggregate data, based on pages 25 to 38 in the Workbook.  The scatterplot program illustrated in these pages fits regression lines to the data and computes the coefficients of the line.  To understand what this means, we first need to understand what it means to plot an equation on a graph.  If we draw two coordinates on a piece of paper or on the whiteboard, we can draw a Cartesian coordinate plane.  with an x-axis (for our independent variable) and a y-axis (for our dependent variable).  We can then plot lines on this graph by using a regression equation:

    Y    =    a   +   b   X.       where X and Y are our variables, and a and b are parameters or fixed numbers given to us by the computer software.

     For example, plot the following lines:
     If a is zero and b is one, then Y = X.  We can say:  if X is 0, Y is 0.  If X is 2, Y is 2, etc.  If we plot these points on the graph we get a straight diagonal line going from the lower left to the upper right (to be demonstrated in class):
      If a is one and b is one, we get a line parallel to the first, but one notch up.
      If a is 0 and b is minut one, the line will go down...   etc.
                                                  
  is a method that computes equations like this to fit straight lines to bivariate relationships between continuous or linear variables.  It works best when the variables are "normally distributed," i.e. when they fit a bell-shaped normal curve with most of the cases near the mean and few extremes. We can see how regression works best by using the scatterplot program in Microcase and the USA data set which has many continuous variables using the US States as the unit of measurement. and clicking on "reg line".  For example, the graph of  % college and Median family income (open Microcase to see this).

At the bottom it says "Line Equation   Y = 15254 + 902.229 X.   This is the equation straight line that appears on the graph. 

What does it mean to say that it is the equation for a line?  It means that if you use the equation to plot points on a graph they will look like that line.  The more general form of this equation is Y = a  + b X  where:
      X is the independent variable  (in this case % college)
       Y is the dependent variable (in this case Med Fam $)
       a is the "intercept" - this is a "parameter" of the equation which means it stays fixed while the variables vary
        b  is the "unstandardized regression coefficient" - it is also a paramater. 
      The software computes the equation for us, which is called "fitting a regression equation to the data".

Sept 9.  By "science" we mean a field of study that attempts to establish generalizations based on empirical observation.  This is different from establishing facts about particular cases as we may do in history or in criminal investigation.  It is also different from mathematics or logic where we try to establish truths through pure reason, or from the creative arts or humanities where we create unique objects of beauty, or from ethics or religion where we reach moral judgments.  There are different ways to divide up knowledge.  Here at Rutgers we have physical science, social science and humanities.  These are broad categories, of course.  There is a long debate about whether social science and physical science are essentially different.  We have not been successful in establishing highly abstract empirical generalizations such as the powerful ones in physics and chemistry.  But we do have a lot of empirical evidence on a more "middle range" level, applicable to certain societies under certain conditions. 
In the social sciences it is often more useful to make policy recommendations than just to state facts, and these recommendations are based on values and moral judgments as well as on empirical data.  We are studying human life and we are part of the systems we study, so it makes sense that we want to make some better.   For example, Florence Nightingale used social research to advocate for better nursing care in the British armed forces during the Boer War.  She invented the bar graph and pie chart.  Felton Earls and his colleagues used a combination of research methods to study the causes of urban crime.  Their organizing concept was "collective efficacy".

So some social scientists believe we should not try to emulate the physical sciences but should take a broader, more humanistic approach.  One way to think about this is in terms of three Greek words used by Aristotle, Episteme, Techne, Phronesis:  Three approaches to knowledge.

Social science begins with concepts as do other fields such as  philosophy and even mathematics if we recognize that numbers are concepts.  The small integers are especially important, especially Zero and One (or nothing and something).  Religion may also start with concepts  The Bible says In the beginning there was the Word, and the Word was with God, and the Word was God.  What does that mean?  The original Greek text uses the word "logos" which means unit of thought or idea or concept, which is where we begin also, with concepts.  We want good words or concepts.  But what is a good concept?  Religious concepts are good if they provoke spiritual reflection, as in reciting a Mantra in Buddhism.  Literary concepts are good if they are beautiful, which social sciences seldom are.   W.H. Auden's poem Under Which Lyre is  an aesthetic attack on social science and other applied sciences.  Social science may not appeal to poets, but it is more useful.  We want concepts that are parsimonious and useful and clearly defined.  We want concepts that help us to make useful discoveries about the observable world.  We want concepts that are falsifiable, which is a key difference between social science and theology or mathematics.   This is an issue now in the debate about "intelligent design" theory, a doctrine that claims to be a scientific theory but many say is a theology in disguise.  Is there any evidence that would disprove this theory.  Is the human body intelligently designed or did it evolve?  Why do we have an appendix?  Why do men have non-functional breasts?  Why are our backs weak like the backs of quadrapeds?  Why do whales have finger bones in their fins? 

In social science we have general ideas or theories, which are statements of relationships between concepts.  From these, we make hypotheses about what we are likely to observe in empirical reality.  We gather data to test our hypotheses, and we change our theories if the tests do not work out.  At least that is how it is supposed to work!  In real life, many social scientists act more like lawyers, selecting facts that support their preconceptions.  We are more successful in being objective in our descriptions than in our explanations or in our predictions.  We know that the  rate has been going down for the last fifteen years or so, but we are not agreed about why. 

The book distinguished "pure" from "applied" and "evaluation" research.  Pure research is motivated entirely by scientific curiosity, applied research seeks to further a goal.  Evaluation research seeks to determine whether a particular program works or not. 

In testing hypotheses, we can make Type One or Type Two errors.  Type One:  accepting a correlation that does not exist.  Type two:  Not accepting a correlation that does in fact exist.  There is a trade-off between the two, to the extent that we avoid making Type One error we increase the risk of Type Two error. 

The null hypothesis is a statement of how things would be if our theory were not true, generally if there was no relationship between our variables.  Some philosophers believe it is more correct to say "we reject our null hypothesis" than to say "we accept our hypothesis as true". 



Sept 7:  We will go through the Introductory Exercise in the Microcase book.  The differences between frequencies, rates and percentages is important.  The frequency is the actual number of cases.  Rates are proportions:  the number of cases divided by a base.  It is important to be clear about the base of the rate.  Rates are often presented as per 1000 or per 10000 or per 100.  If it is "per 100" that is the same thing as a percent.  Also note the difference between aggregate data (data about geographic or other units, in this case states) and survey data (date about individuals).  The "unit of analysis" is the entity that the data describes, e.g., a state, an individual, a family.

Sept 2:  A representative of NJPIRG made a presentation about their programs.  You should contact them if you wish to work on one of them.   We went over the syllabus and class schedule and discussed the use of WEBCT.  To access WEBCT, you use your regular Rutgers username and password.  Most of you have been automatically enrolled in the course.  If you can get into your WEBCT home page and our course is not there, email me so you can be added.  Books are available in the bookstore, it is OK to buy a used book if you can get the software either with the disks that come with the book or by downloading it from http://www.microcase.com/files/CSRM3_Online.exe.