SOAN 2120 Final: EXAM FINAL NOTES

33 Pages
24 Views
Unlock Document

Department
Sociology and Anthropology
Course
SOAN 2120
Professor
David Walters
Semester
Fall

Description
What are the two research methods? Qualitative and Quantitative What are Qualitative methods? They are methods in which we generate the data They are done through interviews and field research They give non numerical data What are Quantitative methods? They are done through survey research, which then you get numerical values Applied statistics is done through quantitative methods What are examples of computer softwares? -Nivivo -SPSS What the two parts of social science? 1. Social theory: encompassed by Karl Marx, Max W Weber and Emile Durkheim. They govern the way in which the world works. They deal with an intellectual part through theoretical perspectives (the way in which we develop our own theories and harnish critical learning. You gain knowledge for the sake of learning. 2. Research: You develop skills for employment (concrete skills) What are some examples of research methods within employment? Qualitative:  Health\hospitals (they want in depth interviews of the patients they are treating)  Private sector businesses (focus groups Experimental Research  Health and hospitals Survey Research -Business\organizations - large scale research 2ND LECTURE What theorist contributed to social thought? -Marx Weber -Karl Marx -Emile Durkheim Who believed in Conflict theory? Karl Marx Who believed in functionalism? Emile Durkheim Who believed in Symbolic Interactionism? Max Weber What is conflict theory? Conflict theory sees society being organized between conflict. There are competing groups (the haves and have nots). There is class struggle. Conflict theorists believe that the power imbalance grows. They believe that there should be equality of result (at the end of the day everyone is equal) What is functionalism? Society exists because they are functional (they serve a purpose). When these functions are no longer possible they come together to find a solution. On the issue of inequality, it is not important that everyone is equal but rather we must provide the same opportunity to those in society Functionalism can be defined by the two following words: consensus and functional What is an example that illustrates the idea of equality of opportunity from a functional perspective vs Conflict perspective? Conflict theorists arge 90% of university professors in the natural sciences are male. This is problematic because 50% of the population is female. In the end, gender distribution should be equal Functionalist: 90% of prison pop is male- their opportunities should be the same What does conflict theory and functionalism have in common? They examine life at the macro level What is symbolic interactionism? This view focuses on the micro level and the individual level. The focus is on symbolism and the meanings that are behind people's behaviour. They want to know why people do what they do and what behaviour is generated. They want to know each individual's perception of reality because it is what governs behaviour What data does conflict theory use? Quantitative and qualitative data What data does functionalism use? Quantitative data? What data does symbolic interactionism use? Qualitative data What are methodologies? How we generate and analyze the data What are the two types of theories? 1. Testable theories 2. Non Testable theories What type of theory does marx have? Non testable What is Marx’s theory about social and economic change? Marx's theory centered around capitalism. He argued that capitalism was the most powerful force on the planet and the darkest. Over time he believed that the rich get richer and the poor get poorer. The brougise owned everything while the proletarian worked at the expense of the wealthy. He thought there would be fewer and fewer rich people that owned most of the resources of the world. He thought that the working class would become education and create a class of themselves (Marx revolution) and they would violently overthrow the elite. They would create a new social system called socialism and it would result in communism. Marx theory is non testable as it is not subject to empirical testing. Psychoanalytic theories cannot be tested What is a theory? A theory is a general assertion about something. You do not test theory directly. You test it indirectly through the hypothesis What makes a theory testable? A hypothesis What is an example of a hypothesis? Guelph students have higher grades on a standardized test than do students from other universities What are the quantitative research methods in testing theories? To recall, quantitative research is data generated through numerical values and this is done through experiments and survey research What are experimental designs? They provide us with a controlled environment and consist of three stages: the pre test, treatment and post test What is an example where you can obtain qualitative data and at the end get numerical values to obtain quantitative data? Testing increases feelings of happiness Hypothesis: group one and two will be happier than group 3 Measure Happiness- video camera- amount of time smiling (qualitative data) There are 25 people in each group and the experiment tracks how long they smile. At the end of this, they get numerical values (quantitative data) What is survey research? Survey research is quantitative research. It does not manipulate the subjects. The information is converted into numerical values in which they analyze. Each question on the survey represents a variable What is a variable? Anything that varies What are the types of variables? *** 1. Quantitative variables 1. Continuous 2. Numerical 3. Ie: education (years of schooling) and income (dollars) 2. Categorical Variables 1. No rank order 2. Non numerical 3. Example: religion What are some examples of quantitative variables? Hours watching television per week, hours worked per week, and age in years What are some examples of categorical variables? Gender (male, female) Field of study (arts, humanities, and business) Province (Ontario, Quebec, Manitoba) What are ordinal level variables? Variables on a continuous scale For example, education Ie: 7. Professional or graduate scalel 6. Bachelor's degree 5. Some university 4. College 3. High school What is likert scale variables? We treat them like quantitative variables. Ie: How happy are you? 4- very happy 3- somewhat happy 2- somewhat happy 1- unhappy When do we treat likert variables quantitative? We treat variables as quantitative as long as there is 4 categories and they are continuous What happened in 1962? Subjects told that psychologists have developed theories regarding how people learn. People learn better if they are punished when they do something wrong. One is assigned the teacher and the other one is the learner. One person is shocked- the teacher will deliver a shock if the learner gets the answer wrong when choosing the answers by pressing the buttons. The early stages of this test past uneventfully. Later on the answers said are incorrect and they get shocks. The shocks increase (more painful). The learner wants to leave and get out, The researcher tells the teacher to continue What were the results? 50% would go to the end and obey the researcher. Before the experiment, the researchers guess 10% What does this experiment show? The forces of obeying someone and having pressure from someone can make us do something we normally would not do What are ordinal variables? They are conceptually continuous thus you can rank them. An example of a ordinal variable is a likert variable which is a scale variable (5 strongly agree….1 strongly disagree). They are quantitative as long as there are 4 categories and they are continuous What is a correlation? The relationship between two variables. We can get data from an experiment and then once they are analyzed by a statistical variable and they are quantitative variable, they are called correlations How do graphs represent correlations? Uphill: left to right→ positive correlation Downhill: right to left→ negative correlation No correlation: flat slope What do scatterplots do? Scatterplots plan an important role when exploring the relationships between two quantitative variables. They allow us to identify a problem causes (unusual observations), the form of the relationship (whether it is linear) and have a preliminary estimated of the strength of the relationship A scatter plot also shows direction and how compact they are together What is a correlation coefficient? A good summary measure of the direction and strength of a linear relationship When does it make sense to use a correlation coefficient? Only when the relationship is linear What does a correlation coefficient range from? It ranges from -1 (perfect negative linear relationship) to +1 (perfect positive linear relationship) 0 means no linear relationship What does +1 mean? Positive slope. One variable increases the other increases What does -1 mean? Negative slope What happens if we get closer to -1 or +1? We get to a straight line What are unusual cases referred to as? They are referred to as outliers. An outlier is someone that is outside the relative area that every other variable is on the graph. This affects the math substantially What do we do with outliers? If it is part of the data but is unusual, we check if there is a coding error When do outliers have less of an effect? If you have a big sample size What is the issue of causality\how do we know which variables causes the other? Correlation does not mean causation. We infer a causal relationship based on our understanding of the temporal relationship between variables. What is the variable that occurs first? Causal variable What is the variable that occurs second? The effect variable What are the classifications of variables? Independent variable→ cause (x) Dependent variables→ effect variable\causes outcome. The dependent variable is influenced by the independent variable (y) What is an example of a dependent, and independent correlation on a graph? Education vs Income Education is the independent variable thus it is on the x axis. Income is the dependent variable this it is on the y axis. A line between the dots means that the experimenters think it is causation. The inference of causality is made by us Why is it sometimes problematic to think that education causes a higher income? Because there may be other variables that are outside the two variables that we are not considering What does a statistically significant relationship mean? It did not happen by chance What does a higher correlation mean? A stronger relationship What does a more steep slope mean? The more steep the slope is and ight the dots are together means that the x (independent variable) is statistically related to the y variable (dependent variable) What is the p value? The likelihood or probability that something will happen What are the common thresholds for “p values? P= 0.5- there is a 5% chance we are mistaken (we are wrong 5 times out of 100) (or 1 in 20) (95% chance we are right P= 0.01- There is a 1% chance we are wrong (1\100=0.01) We are 99% sure we are not mistaken P= 0.001 - It is 1\1000 chance we are wrong- We are % 99.9 sure we are not making a mistake Why do we need p values? Because we can’t sample a whole population thus we need p values to tell us how confident we are in the relationship between the two variables Do statistics equal mathematics? No, statistics deals with probabilities. We never use the word prove because there is no absolute. We can assign an exact p value in which we can be wrong What does a spurious relationship mean? A realtionship influenced by a third lurking variable (ie: instead of education influencing income, it is intelligence that affects both education and income) What is regression analysis? Identifying whether the effect on an independent variable on a dependent variable is statistically significant, while holding other variables constant What are the advantages of survey research? Surveys are very good to tap into attitudes, they are quick and easy to administer, generalizable and can be used on a large number of respondents What are the disadvantages of surveys? Control- we cannot be sure we controlled for everything What does quantitative research generate? Theories What are some attributes about qualitative research? It does not involve statistics but rather sensitive information. It establishes a report and is not restricted by a rigorous research agenda. How is qualitative research done? 1. Qualitative interviews 2. Participant observation What is qualitative interviews? They are open ended questions (ie: tell me how you feel). The respondents answer in their word as they are not restricted. The researchers can probe (rapport) What are the strengths of qualitative interviews? It provides a more, rich, descriptive detailed information. It is better suited for sensitive issues What are the weaknesses of qualitative interviews? It is time consuming, expensive and the effectiveness depends on the research we What is participant observation? It is observing the people in their natural habitat. It is a hands on approach. Researchers participate (rapport). It is based on the principle of grounded theory (the data generates the theory). It is exploratory and here is no variables. In PO, the researcher also conducts interviews What are the strengths of participant observation? It is more interesting, more engaging, more valid, more thorough, and more generalizable (real world) What are the weaknesses of participant observation? It is more subjective (researchers point of view), it is generalizable (beyond the group) What does quantitative data allow? They test theories- they are objective, free from bias, reliable and less thorough What does qualitative data allow? It allows one to generate theories 5TH LECTURE What is an example of a quantitative scenario within the school system? The SOAN exam is quantitative because everyone is evaluated by the same criteria What is an example of a qualitative scenario within the school system? Essay format (you don't know if everyone is measured the same way) Out of experimental designs, survey research, qualitative interviews, participant observation, which is the best? None, they all have their weaknesses. We have to use which is best suited for the purpose What did the Milgram experiment deal with? Obedience What is the ethical issue in this experiment? Did he protect the well being of the participants in this design What are the implications within this experiment? Experimental design: this experiment would not be conducted today as it was a pioneer study What was the Humphries (1970) study? It dealt with impersonal sex in public places in tea rooms. Tea rooms are where gay men would get together Humphrie pretended to be apart of this group. The police would often play the same role as Humphrie as being gay was illegal. He served out as the lookout person to see if anyone was coming. He followed the people to their cars and got their license plates. He went to their house and conducted an interview. His findings were that these people were typical, everyday people. This was surprising to hum because they were engaging in deviant behaviour Why were these findings influential? They had implication on participant observation in the future What was the main issue of this study? Consent- the people in his study did not know they were apart of it What are the dimensions in research ethics? -Privacy -Sensitive questions- values, beliefs and backgrounds -Observation -Informed consent -Deception: it is imperative that we deceive them to a certain degree -Anonymity: the people in the study are anonymous- even the researcher cannot identify them If you can't have anonymity what should you have? Confidentiality- the researcher knows the name of the people of the study, but does not disclose this information What are implications of past research? There is a code of ethics (review boards). At each institution, the wording may be different, however the essential premise is the same. There is a lengthy review process What are some features of ethics in field research? The researcher has to be careful about the researcher\participant relationship as they there is certain illegal behaviour. The law is a tricky area for social researchers Lecture 5 (part 2) What are measures of central tendency? They are statistics that summarize the central location of variables. They are found with quantitative variables What are examples of central tendency? 1. The mean: the average. Add up all the values and divide it by the total number of observations 2. The median: The middle value. You have to organize them from low to high (ie: 3,4,5,5,8,8,8,9=6.5) 3. Mode: The number that occurs the most What are measures of dispersion? 1. Variance: the average square deviation from the mean 2. Standard deviation: the average deviation from the mean What are distribution curves? They represent the distribution of observation across the scope of all potential observations What does a correlation to the left mean? Positively right skewed distribution The mean is greater than the median What does a correlation to the right mean? Right: negatively\right skewed distribution Mean is less than median What does a normal distribution mean? Mean=median. What does a standard deviation from the mean allow us to figure out? The probability What is a z score? It determines if it is statistically significant. 95% of the cases are approximately 2 standard deviation from the mean What z score is significantly significant? A z-score of approximately 2 is statistically significant at 0.5 (95 confidence level) 99% of cases are approximately 2.6 standard deviation from the mean A z score of approximately 2.6 is statistically significant at 0.01 (99% confidence level) How to calculate standard deviation: 1. Calculate the mean 2. For every observation 1. First obs minus mean= value. Square that value 2. Second obs minus mean= mean. Square that value 3. Last obs minus mean= value. Square the value 4. Sum all of the squared values to get a new value (sum of squared deviations) Next divide that value by the sample size (n) minus 1 What is statistics? The science of taking out a portion of numerical data and studying it. It is about using the smaller group of number to tell us things about the big groups of number. In statistics, the statistical population is the large group where you will gather data What is a sample? A randomly picked portion of a population What are the elements of a statistical problem? 1. Identify the question being asked, and identify the population 2. Determine the design of the population (can you use the population or a sample) 3. Establish a method to collect and analyze the data 4. Choose a procedure for making inferences (centers around probability) 5. See how reliable your inferences are 6. Organize data in data sets What are graphical methods? Relative frequency histogram 1. Divides the data into intervals (classes) 2. You can see what proposition falls into each class 3. Normally 5-20 classes 4. There is a relation to the distribution curve What is a class width? How wide classes are (All classes→ equal width) What are class boundaries? Where a class begins and ends What conditions does a class has to meet? 1. Lowest class contains lowest boundaries 2. No measurements on boundaries Steps of making a histogram: -Once you found the range, find classes -Once you determined the number of classes, figure out class width -Class width→ boundaries -Classes→ Take values from data set When are histograms used? With smaller data sets When are distribution curves used? With large amounts of data What is a stem and leaf display? It is a picture representation of data. It shows us the frequency and shape of the data. Unlike the relative frequency histogram, the stem and leaf display allows you to obtain the actual representation of data. The numbers fall into patterns. Each number is divided into two parts: a stem and a leaf. Anything that comes before the decimal point is the stem and the things that come after is the leaf. How do you build a stem and leaf display? To build the stem, you separate and break up the values. Vertically list stems in ascending order (larger on the bottom and smallest on top). Smallest on the left largest on the right. The picture is formed by the numbers of leaves on stem. What does the variant tell you? Tells you the average square deviation around the number LECTURE 7 What are the measures of variability? (shows how spread out the data is) range, deviance, variance What is the range? Highest value- lowest value= range What is the deviation & how do you calculate it? How far away a measurement is from the mean or average set (x-mean= deviation) How to calculate the deviation? Step 1: find the deviation for each value Step 2: square deviations Step 3: find sum of the squares Step 4: divide by #1 of values minus 1 What is Standard deviation? Standard deviation is the positive root of the deviance. What is the variance? variance is the expectation of the squared deviation of a random variable from its mean, What are the 3 measures of central tendency? Mean, mode and median What is the variance formula? It is the same as the standard deviation formula, just without the square root at the end and minus 1 What is the empirical rule? Every measurement is plus or minus 3 deviations from the mean What is an exception to the empirical rule? Outliers are special cases In normal distributions what is the percentage of measurements that will fall within the mean? 68% What are measures of relative standing? The comparison of two variables to each other What are the 3 measures of relative standing? 1. Z scores (see definition below) 2. Quartiles (separate data into 4 tiles - Outliers (can bring data way up or down 3. Percentiles: rank measures by percentage What are z scores? How many standard deviations they are from the mean You take x and subtract by the mean and then divide by the standard deviation 2.0 is the score to be statistically significant How do you test the statistical significance? By testing theories and z scores How can we understand z scores? Every observation can be converted into a z-score. Z scores can indicate whether an observation is staticaly significantly different from the mean. Z scores can be both positive or negative. What does a positive z score mean? It is higher than the mean What does a negative z score mean? It is lower than the mean What are the general rules of z scores? 95% of all observations will be within plus 2 or minus 2 standard deviations of the mean 99% of all observations will be within 2.6 deviations of the mean Thus, if an observation z score is plus or minus 2 standard deviations from the mean p=
More Less

Related notes for SOAN 2120

Log In


OR

Join OneClass

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

Sign up

Join to view


OR

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

So we can recommend you notes for your school.

Reset Password

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

Add your courses

Get notes from the top students in your class.


Submit