Chapter 1
Science, society, and criminological research Process of doing research; conducting research Talk about why we do social science research? o Better understand experiences of others o Try to help live in a diverse society o Learn more about other
March 5, 2008 Conceptualization and measurement Concepts o Concept a mental image that summarizes a set of similar observations, feelings, or ides o Conceptualization the process of specifying what we mean by a term; defining it This is not split
March 3, 2008 Qualitative compared to quantitative data analysis Qualitative Quantitative Focus on meanings Focus on quantifying social phenomena Collection of many data on a few cases Collection of few data on many cases Study in-depth and in detai
Feb 29, 2008 Basic steps in field research o Entering the field o Developing and Maintaining relationships o Sampling people and events o Collecting data Entering the field o Read about the setting o Think about how to dress, how to conduct yoursel
Feb 27, 2008 Qualitative research Three major designs o Participant observation Develop sustained relationship with people while they go about their normal activities o Intensive interviewing Open-ended, relativity unstructured questions Intervie
Feb 25, 2008 Hypothesis o To use deductive reasoning, first state a hypothesis that is derived from propositions presented in a theory o A statement about what you expect your data to show, based on the theory you are testing o We use variable to tes
Feb 20, 2008 The process of doing social science research o Writing research questions o Questions about some aspect of society that you seek to answer by collection and analysis of data Many possible sources Theory Prior research Researchers obs
Feb 18, 2008 Avoids errors in reasoning through use of good research methods I o Reduce likelihood of overgeneralization, when we conclude that what we have observed to be true for some cases is true for all cases o Ex. Study a wide range of school
Feb 15, 2008 Chapter 1 Science, society, and criminological research Process of doing research; conducting research Talk about why we do social science research? o Better understand experiences of others o Try to help live in a diverse society o Lear
May 16, 2008 Multiple Regression Controlling for a third variable o We compare only similar groups o We remove the influence of a second independent variable o We can control for several variables at the same time Multiple regression equation o The
May 14, 2008 Multiple regression Review: Causation o Three general conditions Empirical association Time order Non-spuriousness o The third, non-spuriousness, is important This is why we statistically control for that third factor Controlling f
March 7, 2008 From Operationalization to Levels of Measurement This is important when we start doing statistics! Four levels of measurement o Nominal: no mathematical interpretation or order (e.g. gender, religion, race/ethnicity) o Ordinal : a mat
From Operationalization to Levels of Measurement This is important when we start doing statistics! Four levels of measurement o Nominal: no mathematical interpretation or order (e.g. gender, religion, race/ethnicity) o Ordinal: a mathematical order, but w
From Operationalization to Levels of Measurement This is important when we start doing statistics! Four levels of measurement o Nominal: no mathematical interpretation or order (e.g. gender, religion, race/ethnicity) o Ordinal : a mathematical order, but
ANOVA Analysis of Variance
Testing Hypothesis with multiple group means (ANOVA) Why analysis of variance (ANOVA)? o To compare means across more than 2 groups We have a continuous (interval or ratio level) dependent variable, and more than 2 groups Exactl
Bivariate correlation and regression are used to investigate how a continuous (interval or ratio level) independent variable (X) and continuous dependent variable (Y) are linearly related. With two variables, we can graph their relationship using a scatte
To compare whether two nominal or ordinal variables are related to one another, we display the data with a cross-tab table and compute a Chi-Square statistic. In the cross-tab, each box containing data is a cell, the column and row sub-totals are each cal
When using multiple regression, we regress the dependent variable on several independent variables simultaneously. The regression tells us how each independent variable affects the dependent variable while controlling for all other independent variables.
Review for Multiple Regression
You decide to examine in greater detail what may cause individuals to start offending early in life. You believe that parental criminality and failure at school may be important causes. You gather data from 400 offenders and
I. Alpha: the critical region of a probability distribution We know that with a normal probability distribution, values far from the mean are less likely than values close to the mean. We need to establish a critical region in the ends of the distribution
When we want to compare means of a continuous variable (measured on either an interval or ratio level) across two independent groups, we perform an independent samples t-test. For example, we may be comparing the average incomes across a sample of female
ANOVA
(analysis of variance) tells us whether a continuous (i.e., ratio- or interval-level) dependent variable is related to a nominal or ordinal independent variable. In other words, we have more than two groups of cases, and we want to see if these grou
Survey Research
What is survey research? o Subjects fill out surveys o Advantages: Collect data from many people Collect data from many social settings, on different topics Most popular form of social research o Versatility Can enhance understanding of a
Sampling, causation and study design
Population and sample o Population the entire set of elements (individuals or other entities) o Sample a subset of elements form the population Sampling overview o Two general types of sampling Probability Non-probabi
May 12, 2008 Regression and Correlation Adding an error term o Our prediction helps, but is far from perfect there is always error involved o A different version of the same equation is: y=a+bx+e o This acknowledges the error involved Comparing b
May 9, 2008 Bivariate correlation and regression Correlation o A measure of the strength of the relationship between X and Y o It varies from -1 (a perfect negative relationship) to 1 (a perfect positive relationship) o It tells us how closely to ou
April 9, 2008 Probability distributions Standard scores o Tell us how far from the mean, in standard deviations, each case is o Negatives are below the mean o Postitives are above the mean If Z=0, the score equals the mean If Z=1, the score equals
April 7, 2008 Probability and distributions Probability o The likelihood of something happening; the number of times an event can occur divided by the times is does o The likelihood something will occur o Probability = 13/52 = .25 <- the probability