Experiments & Terms

bohoxadu's version from 2017-10-09 19:31


Question Answer
Experiments rare in sociology - sociology is limited via experiments because people are too complex (impossible)
Experimental groupsdo get the treatment of what’s happening
Independent VariableThe variable that influences another variable without being influenced by that other variable
Dependent Variable A variable that changes in response to the independent variable
Controlled Experimentsyou rarely see them in sociology, they are in a certain setting Ex: a lab you are controlling the environment, its hard to get to answers the same
Field Study more likely to be seen, have to let reality happen, they happen in the field, not hard science. You can get a vague idea of what’s happening
Statistics Mathematical science pertaining to the collection, analysis, interpretation and presentation of data
Mean this is the most common— it means the average. Add up all the numbers and divide by how many there are= 6.6666
Median the middle number once you liked them up from least to greatest—ex: 2,2,4,5,8,10,10,10
ModeMode- the number that is most frequent— ex: 2,2,4,5,8,10,10,10 -- it is 10 here
Causationwhen one variable causes change in another
What are the three things you should look out for when looking at causation?(1) Correlation (2) Temporal Priority (3) No Spurious Correlation
Correlationa relationship between two variables, when one goes up the other goes down
Temporal Priority one thing has to happen before another, one factor has to happen first
No Spurious Correlation when there is a hidden third variable (EX: with TV and planes)
Operational DefinitionThe conversion of abstract ideas into specific, observable things or events.
Variable Anything that can change or that can be sorted into more than one category or value.
When looking at statistics...we tend to hold them up as factual because they are numerical but that is not always true. You have to think of where the number comes from. We tend to think of them as rational.
Reliability The degree to which the results of a study would be repeated in other similar studies.
Researcher BiasThe influence, deliberate or not, a researcher exerts to get the preferred result.
Number Laundering This is when someone’s best guess is repeated again and again as though it is a fact. Usually maybe will be someone well known, but once its stated its taken as fact even though it’s their ball park. They can be poorly worded, or vague and this can be intended or unintended.
Example of Number LaunderingThe number of kids in US killed by guns has doubled since 1950. (If 10 in 1950 died then 20 in 2016) someone took the stat and wrote it out. Every year since 1950 the number of American children has doubled. This phrase implies that every year it doubles. By 1995- that means that 35 trillion children who have died. We have to be careful with how it is phrased. They repeated the statement over and over again as though it was a fact.
People's Agendas & Motives you have to think of who collected the data. We want to show how important data is but it can lead to intentional or unintentional data. Ex: (from a website she showed us) Gun deaths in Florida— hasn’t changed that much, gone up a little, but not that much. The chart is false representing the data the graph is not written correctly (you need zero on it) the data is represented very poorly.
What are the weaknesses of using charts and the data used?- Charts can falsely represent data (zeros can be off) --- Some studies try to make big claims, might be n=4, they only studied 4 people and make a generalizing claim about people. Studies funded by sugar companies show different data than those not funded by sugar companies about sugar leading to weight gain. Also global warming, some scientists made it look worse than it was and then were found out and it hurt the cause. Over exaggeration can backfire. Smoking company funding smoking studies. Data can be manipulated
Presentation of Statistics who is presenting them, how, presented in pop culture (could lose control of it) If someone questions the data they can be accused of questioning the cause itself Have to do critical thinking on what you see
What does "n" represent in n=200? · Represents the population of the study, how many people were involved. · This can be important because some studies try to make big claims ex about race and intelligence and that can be just the amount of people they studied, you can’t generalize all people. (N=4) EX: sugar and weight studies EX: global warming EX: autism
Who collects data and how?data can be manipulative, the representation of staticsistic, who is presenting them and how, it can backfire, but if you question a statistic you are questions the cause itself. How its presented- a lot of what we know is from the media. We have to think critically about what we see
Innumerate don't know how to read numbers, illiterate in that sense
What are the weaknesses of statistics?We have to ask who produced this data, what are the measures, who’s included and who isn’t. What questions were asked, how we are defining certain things, etc. Stats give us important data but we have to think critically about them. Stats can’t tell us WHY- which is a fault. You can’t ask “why”— that is a qualitative question not quantitative
Case Studies Not used often but it is when a researcher focuses on a specific event and occasionally an individual. (EX: natural disasters) (They often come off as biographies)
Secondary AnalysisThe use of available data gathered by another researcher or agency such as the Census Bureau. This works for quantitive not qualitative.
Analysis of DocumentsA lot of times can be written sources Ex: photos, movies can be useful. (Won’t just use documents will often be paired with something else)
Unobtrusive Methods Rarely allowed. You have to be able to study people without them knowing that you are. (Hawthorne- when the people know they are being studied they react differently) EX: it may be watching someone’s garbage etc. Ex: watching people who are interacting with pigeons— but you don’t tell them because if you do it will change their behavior.
Ethics and Research We are dealing with complex human beings; we have to question what we do.
Tuskegee ExperimentThe US government did a study of syphilis on AA’s male population and didn’t treat them. At first there was treatment they were studying what happened to the human body over time. They were receiving a placebo. it started in 1932 (?) it went over 30 years. it went through when we developed medicine to address syphilis even when meds were made they were denied medicine they went to doctors and told them to send them away if they ask for medicine and people died. our government lied to them. there are issues of racism here-the population is refereed to an Ethiopian population. Tt was morally unjustifiable. it was done with the army, and doctors and ended up being a lot more about racism in America then about the actual disease. it is a case where humans were a part of study and were injured and some died while participating in the study.
IRB - Institutional review board/Committees for the protection of human subjects1. every college that received funding has an IRB . 2. The IRB must approve your research before you interact with humans for your research . 3. Privacy, Anonymity, and Confidentiality . 4. Cannot misrepresent yourself
Informed Consent (IRB)the participants must sign a letter that states exactly what you are researching and how your data will be used
What are the risks of IRB?must be informed of the risks - not one can ever be harmed - risks cannot be greater than those of everyday life a) Vulnerable population - children/minors, prisoner, fetuses, pregnant women, cognitively impaired people, etc.
Privacy, Anonymity, and Confidentiality (IRB)should not be able to tell who is being studied specifically, pseudonym, not be specific about characteristics, places too
Misrepresentation (IRB)not misrepresenting yourself (like gov did in syphilis case), except in unobtrusive cases but can not interact with the people, watch from afar

Recent badges