Week 6 notes
Popular in Research Methods in Criminology
verified elite notetaker
Popular in CRIMINOLOGY
verified elite notetaker
This 5 page Class Notes was uploaded by Janaki Padmakumar on Sunday October 9, 2016. The Class Notes belongs to CCJ3701 at University of Florida taught by Chris Gibson in Fall 2016. Since its upload, it has received 3 views. For similar materials see Research Methods in Criminology in CRIMINOLOGY at University of Florida.
Reviews for Week 6 notes
Report this Material
What is Karma?
Karma is the currency of StudySoup.
Date Created: 10/09/16
Week 6 notes Two categories used in social sciences (general categories; specific techniques of each exist) Probability o Objective, random chance sampling o Utilize probability theory for confidence of sampling and generalization o Margin of error in estimations- difference in variation between sample and population for a given variable Non probability o Not up to chance- more subjective, but valuable under certain circumstances (when sampling frame is incomplete) o Snowball sampling- first individual puts you in contact with others to get more participants What is a probability sample and why would we want to draw a probability sample? Population- entire set of elements (unit of analysis) that we are interested in; e.g. individuals in Tampa project Sample- subset of elements from general population (not identical to population) Generalizability- from sample to population (sample generalizability) Sampling error- difference between characteristics of a sample and characteristics from which it was drawn Representative sample- looks like the population from which it was selected in all respects that are potentially relevant to the study; similar distribution without over/under representation How to get a representative sample? Use a method of selection in which only chance determines selection of elements Produce an equal chance of people being selected Probability sampling methods- every element has an equal chance of being selected; known independent chance of being selected. Does not guarantee: Selecting elements of incomplete list (sampling frame- how complete is it relative to population?) Failure to get adequate response rate No response given Size of sample and homogeneity- sharing same characteristics, no large variation in characteristics; as homogeneity increases, sample size decreases due to lack of variation. Heterogeneous populations require larger sample size to accommodate for diversity within the population Error due to chance Two things to reduce random error: o Larger the sample, more confidence held in inferential abilities (sample being more representative of populations chara- little error, more representative= inferential abilities) o More homogenous the sample, the more confidence in its inferential abilities Ultimately say what is level of confidence of estimate, and how much is margin of error Probability sampling- terminology Confidence interval (margin of error) o Plus or minus figures reported in news or tv o Random error o 47% say yes with CI of 4 (range of values) so actually 43-51% support- if asked to entire population Confidence level o Indicates degree of certainty o How often are you willing to be wrong? Need 95-99% confidence in social sciences Often a correlation between confidence interval and sample size; larger sample=less error Statistical tests Making inference from sample to population o How sure can you be that sample results are similar to population o Can result be due to sampling error o Statistical significance testing Probability sampling techniques: o Simple random sampling Elements have equal chance of being selected Done on the basis of chance- generation of numbers or identities Random number tables, lottery procedure, SPSS program etc Equal probability of selection method o Systematic random sampling o Stratified random sampling Proportionate Disproportionate o Cluster sampling Large populations NCVS Random samples of groups or clusters Sample from within clusters Multi stage procedure PHDCN study Causes and pathways in juvenile delinquency- how kids develop within the context of their neighborhoods Adult crimes and substance abuse Measures of how neighborhoods change, and have measures of children growing up over time- representative of social order and children's development IRB- institutional review board Two purposes: o Assess overall risks to human participants o Safeguards evaluated to protect safety, confidentiality and welfare of subjects Bad Blood- Tuskeege syphilis experiment o 600 African American sharecroppers infected with syphilis to examine progression of disease on the body o Investigators recruit participants with free medial care, food and burial service o All participants came from impoverished backgrounds and already had syphilis o Participants told they had "bad blood"- never informed they had syphilis o Started in Macon county 1932-1972 o Penicillin accepted as treatment in 1943 o Penicillin widely available in 1952 o Study exposed in 1972 and participants were given treatment by 1973; treatments extended to families in 1975 Aftermath: o 1997 formal apology issued o 5/8 remaining attended o Damaged trust of black community and made them reluctant to participate in programs o Persistent rumor in black community How did things change? o Most infamous biomedical research study in US history o 1979 Belmont report o Establishment of OHRP- office of human research protections o Led to federal laws and regulations being implemented Ethical issues in research: No harm to participants o Balancing benefits of doing the research against the possibility of harm to subjects or others o Minimize risk and maximize benefit- must apply to both participants and researchers E.g. IRB questions risk of collecting data with interviewers in violent crime neighborhood in Tampa project o Includes psychological and physical health issues Voluntary participation- no coercion o Participation in research must be voluntary o Coercion can be psychological, can be context of the study; no forced participation o Prisoners Anxious to participate most of the time-but might think that by participating, they get favorable judgment, or can end sentence faster By not participating- feel as though they could be punished by CO or warden Need to be clear that not participating has no consequences Anonymity- cannot link identifier to a particular person's data o Protection of participant's ID o Researcher cannot associate a given piece of information with a person Confidentiality o Able to link information with a person's ID, but promise not to do so publically o How will data remain secure? Deception of subjects o Not revealing what the study is about- giving generic topic in informed consent o Concealing ID as a researcher o Debriefing if justification is used Milgram's obedience experiment Psychology study conducted at Yale Study on obedience and humans' response to authority Deceived about nature of experiment and told it was a teacher/learner experiment "Teachers" told to give the "subject" an electrical shock for wrong answers- voltage increased each time per wrong answer Confederate- "undercover" researcher Analysis and reporting Ethical obligation to our colleagues in the scientific community Obligations to make shortcomings of a study known Negative findings should be reported Often report positive findings and neglect others Legal liability Ethical issues expose researchers to criminal liability- collect info on illegal behaviors Observing criminal acts and not reporting Having knowledge that research subjects have committed illegal behavior- self report surveys or field interviews Informed consent- important part of IRB- form signed by participant and researcher that gives overview of project and any associated risks/benefits The Willowbrook Study 1963 to 1966 studies at the Willowbrook state school, a new york state institute fro "mentally defective persons" Parents who cant handle kids who acted out sent their kids to the facility- unaware of abuse being carried out on children Designed to gain an understanding of natural history of hepatitis and subsequently to test the effects of gamma globulin in preventing or ameliorating the disease Deliberately infected children with hepatitis virus Fed extract of stools from infected kids Eventually injected Investigators defended the injection of these children citing that the kids would've contracted the disease anyway- "controlled" environment allowed for monitoring Willowbrook closed its doors to new inmates, claiming overcrowded conditions, kept hepatitis program open for children The Belmont report Issued in april 1979 by national commission for protection of human subjects of biomedical and behavioral research Est by responsibility of investigator to submit research activity for review by IRB Three components Respect for persons- freedom and capacity of subects must be protected, informed consent is central, each subject is autonomous agent Beneficence- all efforts done under principle of "do no harm" maximize benefits and minimize risks Justice- researchers must ask who should receive benefits of research and bear its burdens; each person must share equally in risk/benefit distribution L 10 Causation and obesity Complex chronic disease developing from interactive influences of numerous factors o Social o Behavioral o Metabolic o Genetic (affected by non genetic factors) Correlations are easily interpreted as causation oftentimes- hard to determine complexity of causation Ultimate goal of research is validity Measurement validity- does the measure measure what we think it measures Generalizability- conclusion supported across different groups, settings or events Causal validity- A causes B Direct association and causation