Class Notes (837,548)
Canada (510,312)
LS 221 (8)
Lecture

LS 321 notes.docx

56 Pages
286 Views
Unlock Document

Department
Legal Studies
Course
LS 221
Professor
Owen Gallupe
Semester
Summer

Description
Chapter 1 Research and Theory  Two issues: form of the theory and relationship between data and theory Degree of Abstraction  “Theory” – explanation of observed regularities or patterns  Three common components of a theory: o Definitions – key term in a theory and what they mean o Descriptions – characteristics of the phenomena o Relational statements – connect two or more variables so that knowing the value of one variable conveys information about the other. Deterministic – two variables always go together in a certain way. If the variables are not related in this way, relational statement must be modified Probabilistic – two variables go together with some degree of regularity, relationship is NOT inevitable. So, finding a case that does not fit the pattern does not mean the theory is modified but variables may just not be related in a usual way  Theories of the middle range and grand theories  Theories of middle range: o Limited in scope o Tested directly via empirical evidence  Grand Theories o General and abstract o Structural functionalism, symbolic interactionism critical theory, post-structuralism, feminism o Few direct indications of how to collect evidence and test  Some social scientists reject research that has no direct connection to theory  Relationship between theory and research o Theory may follow upon or arise from collection and analysis of data Deductive and Inductive approaches  Deductive method is most common o Come up with a theory that explains a phenomenon and then deduce hypothesis from it to be tested with empirical data o Confirm or reject the hypothesis o If data does not support hypothesis, theory has to be revised or rejected o Sometimes, a new theory may come to mind during data-gathering stage or relevance of data for a second theory is apparent after collecting the data  Inductive approach o Theory is the outcome of the research o Researcher first gathers or examines data relevant to phenomenon first. o Data is gathered not to test a theory but to come up with the information required to construct a theory. o Data gathering comes first, create concepts and theories after o Strategy of moving back and forth between data and theory is iterative. o Practice of deriving theories from qualitative data – grounded theory o Although some researchers use induction to develop theories, sometimes, the results are more than empirical generalizations. Epistemological considerations  Notions of what an be known and how knowledge can be acquired  Should social sciences follow the same principles and procedures as natural sciences? Positivism  Only phenomena confirmed by the senses is accepted as knowledge – principles of empiricism  Ideas must be subjected to empirical testing before being considered knowledge  Generate hypothesis that can be tested and allow explanations of observed laws and principles.  Knowledge can be arrived at through gathering of facts that provides the basis of laws – induction  Science must be value free – conducted in a way that different researchers with same data will reach same conclusion  Clear distinction between scientific statements – how and why certain phenomena operate the way they do – and normative statements – certain acts or social conditions are morally acceptable  Includes elements of both induction and deduction  Possible to collect observations without reference to any pre- existing theories and develop new theories on the basis of those observations  Not synonymous with science and the scientific Interpretivism  Role of social scientists to grasp the subjective meanings of people‟s actions  People act on the basis of meanings that they attribute to their acts and those of others  Believe that it is the job of the social scientist to gain access to the common thinking of the people they study and interpret people‟s actions from the point of view of others  Symbolic interactionism is a sociological perspective that falls under interpretivism. o Individuals self concept emerges through an appreciation of the perception of others  When social scientists adopt an interpretive stance, they aim to place those interpretations into a social scientific framework Ontological considerations  Two ontological debates o 1. Do social phenomena have an objective reality? If yes, you are in the objectivist camp – no such thing as a social reality o Is what passes for reality a set of mental constructions? If yes, you agree with the constructionist position – no facts, only interpretations and no objective social reality o Soft constructionist position – there may be an objective social reality but many of our ideas do not reflect it – those ideas are constructed to justify or rationalize various forms of domination o 2. Is social reality akin to the physical world as people see it – individuals and group have to confront but have little control over? If yes, you support variant objectivism o Is social reality not necessarily pre-existing and fixed but rather created through our actions? – If yes, agree to constructionism o Constructionist perspective  Concepts people employ to help them understand the natural and social world are social products whose meaning is constructed in and through social interaction Relationship to Social Research  Ontological assumptions and commitments affect both the way research questions are formulated and the way research is carried out General orientations: quantitative and qualitative  Quantitative – numbers and statistics, formal and mathematical measurement and analysis techniques o Deductive approach to the relationship between theory and research where theory testing is a prime objective o Incorporates the practices and norms of natural science and positivism o Embodies a view of society as an external, objective reality  Qualitative – words and symbols o Inductive approach to relationship between theory and research – generation of theories and intepretations is main goal o Rejects natural science and positivism and replaces with methods that determine how individuals interpret their social world o Social reality is a constantly shift and emergent property of individual creations o Sometimes, qualitative research is used to test rather than generate theories Influences on the conduct of social research Values  Values are a form of preconception and they should be suppressed when conducting research  Social researchers are always influenced by value presuppositions that have implications for the conduct of social research  Researchers may develop affection or sympathy for the people they are studying  Values can intrude at any points in the social research process including: o Choice of research area o Choice of method o Formulation of research question o Formulation of research design and data collection technique o Data collection o Analysis of data o Interpretation of data o Conclusions  Dealing with values and bias – research is not value-free and research process is made explicit  Influences on social research o Theory o Values o Politics o Ontology o Epistemology o Practical considerations Politics in Social Research  Social researchers sometime take sides  Research funding – social research is funded by organizations like private firms or government departments that may have an interest in the outcomes – only invest in studies useful to them o Research funded by the government is typically empirical and quantitative o Political issues usually arise when the funding agency itself is trying to secure a continuous stream of government funding  Access to organization is mediated by gatekeepers concerned about the researcher‟s motives but also about what the organization gains from the investigation  Gatekeepers seek to influence how the investigation will take place, what questions will be asked and who can and cannot be a focus of study.  Public institutions and commercial firms are concerned with how they are going to be represented in publications.  Gaining access is almost always a matter of negotiation and turns into a political process. Practical Considerations  Choice of research orientation, design or method has to match the specific research question being investigated  If a researcher is interested in a topic with little research, a quantitative strategy may be difficult to employ  Qualitative exploratory approach may be preferable because that is associated with the generation of theory rather than theory testing  If a researcher is interested in a topic where little or no research has been done, a quantitative strategy may be difficult  Choosing a research question o Purpose of the study in the form of a question o Research often starts with the choice of a general area of interest o No single study can answer all the research questions o Research questions in quantitative research are more specific than qualitative research o They set realistic boundaries for research and are crucial because they guide  Literature search  Decisions about the kind of research design to employ  Decisions about what data to collect and from whom  Analysis of data  Writing up the findings Chapter 2 Research Designs  Nomothetic explanations have to satisfy three criteria of causation:  Interested in cause and effect, general laws and principles.  Attempt to generalize to broader population o e.g., „The prevalence of suicide in a particular social group is a function of the level of integration individuals typically have in the group.‟ o They look for principles like x causes y. For example the Durkiems theory is that one thing influences the other. o Correlation – cause and effect have to vary together o Time order – proposed cause must precede the effect in time o Non-spuriousness – alternate explanations for correlation observed have to be ruled out  Spuriousness - False or illegitimate o Quantitative researchers choose research designs that satisfy these criteria o Qualitative researchers produce a rich description of a person or group based on the perceptions and feelings of the people studied rather than to discover laws and principles o Once a design has been selected, a specific method for collecting data has to be chosen  Preset instrument: questionnaire, interview schedule, participant observation  Ideographic explanations are based on rich descriptions of a person or a group o Not meant to apply to persons or groups who were not part of the study o e.g., „Jade became addicted to crack because she never got over her parents‟ divorce, she felt she was never really accepted by her friends, and had a classmate who offered her crack,‟ plus many more details of how Jade interpreted her life o Not concerned with laws but rather what goes on in a particular group and getting know how people experience certain events. When you get to this level of detail, generalisation is much harder to make Criteria for evaluating Social Research  Variable is a characteristic or attribute that varies  Three of the most prominent criteria are reliability, replicability, validity Reliability  Quantitative researchers are concerned with whether measures of social science concepts are reliable  Results should remain the same each time a particular measurement technique is used on the same subject  Would the same results be achieved if the measurement technique were administered many times to the same subject? Replicability  Replicable if others are able to repeat it and get the same results  The initial research must spell out all the research procedures in detail for replication to be possible.  The results remain the same when others repeat all or part of a study Validity  Integrity of conclusions generated by a piece of research  Three main types of validity o Measurement or construct validity – quantitative research and it‟s measures of social concepts  Refers to whether a particular indicator measures what it is supposed to measure  “Are you measuring what you want to measure?”  e.g., Is education a valid measure of socioeconomic status? o Internal validity – relates to casualty  Can the study establish causal ordering? Do we know which is the cause and which is the effect?  e.g., did the study establish that personal income level in Canada really is influenced by one‟s level of education? Could income be influenced by something else?  Proposed causes is independent variables and corresponding effect is dependent variable.  Independent variable – predictor, occurs first. Eg: level of education  Dependent variable – outcome, occurs as a result of independent variable. Eg: level of income  Does the independent variable really have an impact on the dependent variable? o External validity – is a study‟s findings applicable in settings outside the research environment  The more the social scientist intervenes in natural settings, the greater the chance that the findings will be invalid  Are the findings applicable to situations outside the research environment? – Naturalistic studies  Can the findings be generalized beyond the people or cases studied?  Studies using representative samples  Asking for grades in one class, doesn‟t mean everyone in UW will have the same grades. o Qualitative research fulfills this requirement of external validity because it takes a naturalistic stance o Naturalism – style of research that minimizes the use of artificial methods of data collection o If a representative sample of people is selected, the researcher can be confident that the results may be applied to the population from which the sample was drawn. Relationship with the General Research Orientation  Internal validity is more pertinent to quantitative research  External validity has relevance to qualitative and quantitative work but representativeness of samples have a more obvious application  Each aspect of trustworthiness has a parallel with the criteria o Credibility – parallels measurement and internal validity o Transferability – parallels external validity o Dependability – parallels reliability o Confirmability – parallels replicability Research Designs Experimental Design  To conduct a true experiment, do something to people and observe the effects  Experiment manipulates an independent variable to determine its influence on a dependent variable.  Dependent variable is observed and measured  Experiments are uncommon in sociology. Manipulation  Experiment manipulates an independent variable to determine it‟s influence on the dependent variable.  Subjects are allocated to a treatment group in which the independent variable is manipulated; others are placed in a control group where no manipulation takes place.  Control variable cannot be manipulated because they may be impossible to carry out and ethical concerns preclude them  May do the things of interest to sociologists like gender roles, political preferences have long-term causes that cannot be easily simulated in experiments.  Laboratory experiments take place in artificial settings o They control research environment o Enhance internal validity because it is easier to assign random research subjects o Easy to replicate  Field experiments take place in real-life surrounds like classrooms and factories Classic Experimental Design  Rosenthal and Jacobson study o subjects are randomly assigned to two groups o experimental manipulation (different levels of independent variable) is carried out on the experimental group o Control group is not given treatment o Dependent variable is measured before the manipulation to make sure the groups are equal at start o If equal (since it is random assignment (participants are placed in the experimental or control group using a random method), researchers feel confident that differences after manipulation will be due to treatment o The dependent (outcome) variable is observed or measured (post-test) in each of the groups at time 2. Threats to internal validity in experiments  Lack random assignment and/or the presence of a control group.  History: some event occurring after the treatment was given may have influenced the dependent variable  Testing; subjects may become more experienced at a test or sensitized to the experiment as a result of the pre-test influencing the dependent variable  Instrumentation: changes in the way a test is administered can account for an increase in scores between a pre-test and post-test  Mortality: participants leave the experiment before it is over  Maturation: participants change over time, e.g., get older, develop mentally and emotionally, etc.  Selection: post-test differences between the control and experimental groups may have been caused by pre-existing differences Threats to external validity  Generalizability of findings derived from the experiment  Interaction of selection and treatment: to what social and psychological groups can finding be generalized?  Interaction of setting and treatment – Can the results of a study be applied to other settings? Can expectation effects be discerned in non-educative settings?  Interaction of history and treatment – can the findings be generalized to the past and future?  Interaction effects of pre-testing – subjects may become sensitized to the experimental treatment and their responses are affected as they become more test-wise  Reactive effects of experimental arrangements – People are frequently aware that they are participating in an experiment and their awareness may influence how they respond to the experimental treatment. Quasi- experiments  Harder to establish internal validity  Natural experiments – experiment like conditions are produced by naturally occurring phenomena or changes brought about by people not doing research o Impossible to randomly assign subjects to experimental and control groups o Groups may not have been equal n all characteristics before independent variable was introduced o Two different social movements, Social Credit and the Cooperative Commonwealth Federation (CCF) came to power in Alberta and Saskatchewan respectively.  Researchers could compare the effects the different movements had on the their respective provinces.  A drawback that researchers of the two movements had to face was that the two provinces may have been different in important ways before the two movements came to power  Those differences rather than rule by two different movements, that produced the differences between the two provinces that developed later.  Threat to internal validity o Prominent in evaluation research studies which examine the effects of organizational innovations like longer school day or greater worker autonomy in plants o Used to evaluate the effectiveness of institutional policies Cross-Sectional Design  Taking observations at one point in time (no before and after comparisons).  They do not include a manipulation of the independent variable (no „treatment‟ is given).  Examples: questionnaires and structured interviewing and structured observation and analysis of statistics or diaries  Collection of quantitative data on more than one case  Researchers are interested in variation between people, families, states – can only be determined if more than one case is examined  Internal validity problems: Difficult to show cause and effect because independent and dependent variables are measured simultaneously o e.g., a researcher may find a positive association between self-esteem and income. o But does self-esteem influence the level of income, or is it the other way around? o Is there reciprocal causation: self-esteem influences income and income influences self-esteem?  If you‟re only asking people about some thing at one point in time, we can‟t deduce cause and effect.  External validity problems: a random method needs to be used to select participants for the study or else, the findings may not hold for people who were not studied  Strength: Cross-sectional research can examine the effect of variables that cannot be manipulated. o age, gender, ethnicity, culture, social class, etc  Issues in reliability - quality of the measures employed to study concepts that the researcher is interested in rather than to research a design  Not possible to manipulate the variables – reason why quantitative researchers use cross-sectional instead of experimental  Collects data on a series of variables for different cases Longitudinal Designs  Cases are examined and repeated over time  Provide information about the time-order of changes in certain variables  Establishes the direction of causation o Eg: if an increase in income is observed at T1, and an increase in life satisfaction occurs at T2, that is evidence that the increase in life satisfaction was preceded by the increase in income, rather than the other way around.  What we think is the cause came before what we think is the effect.  No manipulation of independent variables – gives insight into tie order of variables  Panel Study: o the same people, households, organizations, etc. are studied at different times – can be randomly selected  Cohort Study: o people sharing the same experience are studied at different times, but different people may be studied at each time o Ex: a researcher may study people born in 1990 at three different times (say in 1995, 2000, and 2005), but may use different subjects each time o Can have panel cohort studies – measure a particular group of the same people repeatedly. o Some people use „panel‟ and „cohort‟ interchangeably – synonymous with longitudinal  Both types of study have similar design structure and are concerned with improving the understanding of causal influences  Problems: o sample attrition – some subjects may die or move away or withdraw at later stages; o Difficult to determine when subsequent waves of the study should be conducted. o Panel conditioning: people‟s attitudes and behaviours may change as a result of participating in a panel. o Can be difficult to concretely determine causality Case Study Design  In-depth study of a single case  Single case - person, family, organization, event, etc  Qualitative and/or quantitative research methods.  External validity of case studies can be an issue.  The findings for a particular case may not be applicable to other cases o e.g., a study of student activism at York university may not be generalizable to students at other universities. o Strengths of case studies: Provide in-depth descriptions that cannot be achieved using other methods.  Types of cases: o Critical: specifically chosen because it is thought to be particularly informative (either supporting or against theoretical expectations). o e.g., studying a high-level delinquent who reformed after participation in a boot camp program. o Extreme (or unique): illustrates unusual cases, which help in understanding the more common ones. o e.g., studying the life of a person who has been married seven times helps researchers understand more common marriage patterns o Revelatory: examines a case or context never before studied. o e.g., the study of a particular historical figure may be enhanced when documents are „de-classified‟ or enter the public domain, such as the diaries of former Prime Minister McKenzie-King.  Chapter 11 Ethical Issues in Research  Step 1: ensure that the people being studied are not harmed  One person/group does not bear an undue burden of the research risk  Knowledge should be of secondary interest  There needs to be a balance between potential gain and risk of harm  If there is a risk or harm, always weighted in favour of research subjects Policies in Canada  Canadian Charter of Rights and Freedoms  Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans outlines respect for persons, concern for welfare and justice  Administered through the Research Ethics Board (REP) o All Canadian research requires approval from the REP before research commences  Policy is meant to prevent inadvertent harm caused to participants.  The researcher may not recognize all the risks. Respect for Persons  Individuals have basic human rights that include dignified treatment by researchers.  Respect for persons is the most fundamental principle  Most quantitative studies and some qualitative studies, participants are given an information sheet outlining what the research is about. o the research project o the methods o the potential risks and potential benefits of participating o the names of those in charge o the affiliated institution and contact information o assurances of confidentiality o how the data will be stored o how the findings will be published  If they have all the information and are willing to participate, they have to sign a consent form which has a brief synopsis of the study and states that research subjects are free to leave at any time without penalty  Consent must be freely given and cannot be coerced. o If there is Limited ability (e.g., child, medically limited), must have an agent or guardian provide the consent.  There has to be free, informed, and ongoing consent  Must be advised of the risks and potential benefits of the research.  If participants were fully informed of the purpose of the experiment, it would be difficult to avoid reactive effects that occur when people know they are being observed so deception might be involved o Deception is a last resort o Use it sparingly o Subjects should be debriefed about the deception Concern for Welfare  Researcher must consider the well-being of person/ group/community o Avoid harm, embarrassment, inadvertent identification  Respect for privacy and confidentiality – research subjects should be identifiable when the findings are published  Subjects are usually identified by code numbers rather than names when the data are processed  Random response technique – designed for interview situations where respondent is asked about controversial or illegal activities o Respondents flip a coin but not reveal the results o Respondents are then instructed to answer the question they are asked carefully  Pseudonyms may not be enough. o Detailed description of physical and social settings in which qualitative researchers provide needed context can offer enough clues to identify the participants  In-depth research, analysis, and reporting may provide enough detail that people or locations could be identified.  Covert research o Very intrusive o No consent o Benefits must outweigh the any harm to participants o Little to no concern over reactivity due to participants knowing they are part of a study o Researchers must ensure anonymity of the non- consenting participants Justice  No person or group should be exploited for research (e.g., prison inmates).  No person or group should be systemically excluded from the benefits.  Principle of no harm should be followed o Less risk in social research than in medical research but still exists (e.g., Zimbardo).  In practice, potential harm should not outweigh the potential benefits. o Avoid research that is likely to cause harm or be disturbing. o Life should not get worse because of participation in a research project. o If potential harm and risks are greater than risks of everyday life then informed consent is a minimum requirement. o Ability to opt-out at anytime  Paid participation o May induce risks beyond participants normal behaviour. o May be viewed as pressure to participate or remain in a study beyond participants usual stopping point.  Ex: It can be bad if you pay $100,000 to sleep with a prostitute who might have AIDS.  It can be okay if you pay an undergrad $20 for an hour of study. Chapter 3 Steps in Quantitative Research  Theory  Hypothesis  Research design  Devise measures of concepts  Select research site  select research subjects/respondents  administer research instruments  process data  analyze data  findings/conclusions  write up findings/conclusions  Theory and Hypothesis: Idea Stage o Use the research method that works for your question  Research design, devise measures of concepts, select research site, research subjects: Design stage  Deductive approach since it starts with theory. Hypothesis is deduced from a theory and is tested.  Selecting a research design has validity issues – external validity of the findings and ability to assign causality.  Operationalization – refers to the operations performed to measure temperature or velocity.  Step 7 is administration of research instruments – pre-testing subjects, manipulating the independent variable and post- testing  Cross-sectional research – interviewing members of the sample with a structured interview schedule or distributing a questionnaire  Structured observation – watching the setting of interest and the people in it and recording the types of behaviors  Step 8 – information must be systematically organized to analyze it.  Processing data entails coding the information which is step 9  Step 10 – interpret the results of the analysis where the findings emerge  Research must be written up in Step 11 – convince readers that the research conclusions are important and a significant part of doing research is convincing others of the relevance and validity  Process of deduction and induction (Feedback loop) is a positivist foundation of quantitative research Concepts and their measurement What is a Concept?  Ideas or mental representations of things – measure of intelligence.  Concept can be independent or dependent variable  Same concept is usually independent in one context and dependent in another Why Measure?  Measurement allows for delineation of fine differences between people in terms of the characteristic in question  Provides a consistent device for gauging distinctions – measure should general consistency results unless the phenomenon being measured has changed  Provides the basis for estimates of the nature and strength of the relationship between concepts  Measurement allows us to see small differences between people on the concepts we are interested in  Allows us to estimate the relationship between concepts and the strength of the relationship Indicators  Nominal definition – describes in words what the concept means like a dictionary o Political party reference: party that people most commonly identify with.  Operational definition – operations the researcher will perform to measure the concept. Most important in quantitative analysis o Definition is contained within the different categories o Political party reference: Do you support the Conservatives, Liberals, Bloc Quebecois etc. o Gives a specific breakdown as to how the concept is measured  Indicator refers to a measure in the ordinary sense  Devising indicators: o Questions that are part of a structured interview or questionnaire schedule o Develop criteria for classifying observed behavior o Official statistics o Developing classification schemes to analyze content of written material  It is advantageous to use more than one indicator of a concept because it: o Reduces the likelihood of misclassifying some people because the language of a question is misunderstood o Ensures the definition of the underlying concept is understood correctly o Allows us to examine various dimensions of a concept o Allows factor analysis (sorting out the degree to which a particular concept is being measured and by which indicator)  By asking question, you can determine if you are measuring different things or all the same things. Using Multiple item Measures in Survey Research  Single indicator may misclassify some individuals if the wording leads to misunderstanding of its meaning  Single broad, general indicator may not capture all the meaning in the underlying concept  Specific question may cover only one dimension of a concept  Multiple indicators allow for fine distinctions and for sophisticated data analysis like factor analysis. Dimensions of Concepts  Interest may have multiple facets or dimensions  People scoring high on one dimension of the concept may not score high on another dimensions  Tendency to rely on a single indicator for each concept Reliability and Measurement Validity Reliability  Consistency of measures Stability over time –  whether the results of a measure fluctuate as time progresses  You get the same result after using a measure at different times assuming no change has happened  If there is no change, the result will be the same o Test-retest method – administering a test or measure on one occasion and re-administering it to the same sample on another occasion. There should be a very high correlation on the responses barring anything that has changed o Those who score high on the first observation should also score high on the second and those who score low on the first should score low on the second as well. o Problems:  Factors can come into play to change these responses  The pre-test may influence a person‟s influences later on.  Respondents answer at T1 should influence how they reply at T2 Internal reliability (internal consistency)  Multiple measures that are administered in one sitting are consistent – if respondents scores on one indicator is related to their scores on other indicators  Multiple measures or different questions are administered in one sitting consistent with each other. Ask multiple questions measuring the same broad pattern or general construct – then there is internal consistency o Eg: How satisfied are you in general? How happy are you with your family life? With you schooling? Occupation? Schooling? – these may or may not measure the same construct  Cronbach‟s alpha coefficient – common test for internal reliability. This test tells you whether the questions are telling you the same construct o Value varies from 1 to 0. 0.7 or higher is considered good  Split-half method – six indicators would be divided into two halves of three allocated on an odd-numbered item.  Degree of correlation between the scores on the two halves who be calculated.  Perfect positive correlation – complete internal consistency would yield a correlation coefficient of 1; no correlation and no internal consistency would produce correlation coefficient of 0 Inter-observer consistency (inter-rater reliability)  If answers to open ended questions have to be surprising, the reliability and validity are matters of concern  When researchers are determining scores on particular concepts on behalf of the respondents Measurement Validity  Are you measuring what you want to measure? o Face validity  At first glance, the measure appears to be valid. Ask for the opinion of someone with expertise in the area of study. o Concurrent/criterion Validity  Is the idea that what you are measuring matches up with other constructs that get at the same idea. Established if the measure correlates with some criterion relevant to the concept. Lack of correlation brings doubt to the validity (correctness)  Ex: people with higher IQ should get better grades so IQ scores Construct validity  Established if the concepts relate to each other in a way that is consistent with the researcher‟s theory  Results match what would be predicted given the theory.  Ex: If a theory states that child hyperactivity increases as parenting gets worse, gives us an idea that measures used to gauge parenting and hyperactivity are valid. o Convergent validity  Measure of a concept correlates with a second measure of the concept that uses a different measurement technique  Measuring things in different ways gives you the same results. What we know is not based on how we ask the question or how we measure but rather, it is valid.  Don‟t want different methods to give different results  A measure that is not reliable is not valid. o Inconsistency in data makes the data unusable. o You can have measures that are NOT valid but IS reliable o Ex: A clock is very reliable but not a measure of life satisfaction Main Goals of Quantitative Researchers  Establish causalty whether A causes B  Do areas of high levels of discrimination have higher levels of robbery? Does discrimination cause robbery?  Are finding applicable to other concepts  The goal is to come up with law like findings that apply to large numbers of people to be able to generalize  This is of particular concern for researchers using cross- sectional and longitudinal design  Having a representative sample is essential for generalization. o Accounts for characteristics for population in general. Taking a random sample which is large enough – any little differences between sample and population are evened out. o Selecting people at random – it should represent population in terms of ethnicity, gender, age, education level, income  Representative sample o Studies only represent the population from which it was taken but if done well, we can attempt to generalize more broadly if the characteristics of a broader population are taken into consideration o Survey Waterloo students but get a generalized sample to the whole city regarding their opinion on political party  Mail surveys  Door to door  Probability Sample o The use of a random sample drawn from a given population  Replication or repeating a study uses the same methods o Provides a check for biases and routine errors o If the findings are not the same as those of the original study, comparison provides reason re-evaluate the methods and findings o Different researchers read the original study and do something similar with a different population. Similar to stability but applies it to different groups of people Criticisms of Quantitative Research  People have subjectivity and agency and they can react to their social environment.  How to come up with laws because of the possibility of subjectivity  Researchers will say that they can study decision making processes by asking people to respond to questions like how they would react to situations o Laws don‟t always apply to everyone  Quantitative research is usually very artificial that we cannot learn anything from it and there is a false sense of precision and accuracy o Problem can arise if people interpret the same survey item differently o The way the numbers are generated is in an artificial way that it does not relate to the real world.  Quantitative research produces a disjuncture between real life and research o Artificiality – can we take quantitative findings and project that onto real life? o Difference between what people say to researchers and what they actually do o This can undermine quantitative survey research  Manipulated independent variables are small facets of life and have short term effects. o One specific aspect can influence an outcome. How do we know that A causes B? o Control for other factors is not there. o Any given social outcome is a product of many different factors that may not be captured by experiments  Explanations for findings may not address the factual perceptions of research participants o Might not represent how research participants actually feel and interpret what goes on o Moving towards trying to come up with quantitative approaches to measure and take into account these interpretations o Valid quantitative findings may be achieved but conclusions do not take into account what the people involved actually think o Questionnaires measure relationships assumed to be important  Might miss important dynamics that researchers have not considered  Quantitative researchers tend to be objectivist o Social reality exists that is independent of the observer and that the social order is fixed o Whether or not there is an objective social reality is debatable o Reality is interpreted – interpretivist approach Quantitative Research: Reality and Practice  Ideals of quantitative research and how it is actually conducted may be very different  Not much focus on reliability and validity  Research is expensive – takes a lot of resources  If you want to be a researcher, you need to put in the effort o Time, cost and feasibility limits what can be done in any research project Establishing Causality  Quantitative research involves a search for causal explanations  Quantitative researchers are rarely satisfied with describing how things are – keen to find out why things are the way they are  Researchers examining prejudice describe but also explain it – finding causes  When experimental design is employed, the independent variable is the one that is manipulated and there is little ambiguity about the direction of causal influence  Research of the experimental type Chapter 12 Key Terms  Element or Unit: single case in the population. If you are doing a quantitative study, it is one row in a data set (people, countries – small groups)  Population: where the elements are drawn from. All cases that a researcher is interested in.  Sampling frame: the list of elements that the sample will be selected from  Sample: the elements (subset of a population) selected for an investigation. The group that you participates in the study. You draw the sample from the population  Representative sample: a sample that contains the same essential characteristics as the population.  Probability sample: a sample selected using a random process so that each element in the population has a likelihood of being selected  Non-probability sample: sample selected using a non-random method  Sampling error: difference between characteristics of the sample and population  Non-response: when an element selected for the sample does not supply the required data  Census: data that comes from an attempt to collect information from all elements in the population Introduction  Who you select to participate is very important in terms of being able to answer your research question.  Sampling is how you go about choosing people to participate  There are different sampling methods but the problem with them are the biases.  Three sources of bias: o Not using a random method to pick the sample – risk that the selection process is affected by human judgement o Sampling frame or list of potential subjects is inadequate or incorrect and sampling frame excludes some cases or the sample may not represent the population even if a random sampling method is employed. o Some people in the sample refuse to participate or cannot be contacted and there is non-response. – Those who participate my differ from those who refuse or cannot be reached  Non-response: People that respond may be different in important ways than people who don‟t respond. Certain groups do not respond in the same way as others  Eg: If you are interested in the experiences of single parent families, the poorest might be less able to participate since they may need to work multiple job Sampling Error  Possibility of sampling error cannot be completely eliminated even with a well-crafted probability sample  Discrepancy between the sample group and the total population  It is using random samples and making the sample as large as possible to minimize sample error  Probability sampling allows you to employ tests of statistical significance that permits you to make inferences about the population from which the sample is selected Types of Probability Sample Simple Random Sample  Best type of sample to use and does the best job in minimizing sampling error because every element within the population has an equal chance of being selected  Sampling ratio is expressed as n/N where n=sample size and N=population size  Steps in devising simple random sample: o 1. Define the population. (N) o 2. Select or devise a sampling frame - a list of elements in the population o 3. Decide on the sample size (n) from the total population (N) o 4. List all the students in the population and assign them consecutive numbers starting at 1-N o 5. Using a table of random numbers or computer program to generate a list of n random numbers from 1-N o 6. The students who match the n random numbers constitute the sample. The sample will be comprised of the cases whose element numbers match the randomly generated numbers.  Almost no opportunity for bias since selection is random  Selection does not depend on availability  Statistic tables have five digit numbers but if population is only four digits, the only numbers that might be appropriate are those that begin with zeros or else drop the first digit.  If the four digit number is still bigger than the population size, ignore it and move to the next one  Without replacement method – no number is put back in for a second chance at inclusio
More Less

Related notes for LS 221

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