Notes classes 1 through 12
Popular in Scientific Methods in social Work
verified elite notetaker
verified elite notetaker
verified elite notetaker
One Day of Notes
verified elite notetaker
verified elite notetaker
verified elite notetaker
Popular in Department
This 75 page Bundle was uploaded by Brandy Loveland on Tuesday November 11, 2014. The Bundle belongs to 510 at a university taught by a professor in Fall. Since its upload, it has received 58 views.
Reviews for Notes classes 1 through 12
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 11/11/14
SW 510Scientific Methods in Social Work Overview Research Questions and Hypotheses Systems of Logic Deductive inductive Research Methods Qualitative quantitative Philosophy and theory Identifying Research Questions Practical problems in social welfare Decisions that confront social service agencies Information needs related to policy planning and practice How to Choose a Topic Think about issues or problems in an agency your field placement where you work or volunteer Talk to people related to the agency staff clients to help you formulate potential research topics Refining Research Questions A problem of manageable size Narrowing down Evaluating Research Questions 1 Feasibility 2 Social importance 3 Scientific relevance Systems of Logic Deductive Reasoning Wheel of ScienceResearch Cycle Terms Functions of Theories in Social Work Research Make sense of observed patterns Shape and direct research efforts Guide intervention amp policy Developing Hypotheses Hypothesis A testable statement of presumed relationships between two or more conceptsvariables Specifies how changes in one variable will be associated with changes in another variable Types of Variables Independent variable precedes causes or explains a change in another variable Types of Relationship Positive or Negative or Curvilinear Developing a Proper Hypothesis Include IV the postulated explanatory variable and DV the variable being explained Specify nature and direction of relationship State so they can be verified or refuted Has clear significance for guiding social welfare policy or social work practice Good hypotheses should be clear specific More than one answer testable and valuefree Conceptualization The process through which we specify precisely what we will mean when we use particular terms specific and unambiguous Conceptual Explication Variables specific concepts or theoretical constructs you are investigating Attributes O O Concepts that make up a broader concept are called attributes descriptive characteristics or qualities of variables manwoman vs gender Operationalization 99 O 9 Translation of variables concepts of interest into observable term The final specification of how we would recognize the different attributes of a given variable in the real world Philosophy and Theory in Science and Research Paradigm a fundamental model or scheme that organizes our view of something Informs our approach to doing research I Positivism Emphasizes objectivity precision and generalizability in research I Interpretivism Indepth subjective understanding of people s lives Quantitative Qualitative Quantitative Qualitative ScienceResearch Cycle Terms Functions of Theories in Social Work Research Make sense of observed patterns Shape and direct research efforts Guide intervention amp policy Developing Hypotheses Hypothesis A testable statement of presumed relationships between two or more conceptsvariables Specifies how changes in one variable will be associated with changes in another variable Types of Variables Independent variable precedes causes or explains a change in another variable Types of Relationship Positive or Negative or Curvilinear Developing a Proper Hypothesis Include IV the postulated explanatory variable and DV the variable being explained Specify nature and direction of relationship State so they can be verified or refuted Has clear significance for guiding social welfare policy or social work practice Good hypotheses should be clear specific More than one answer testable and valuefree Conceptualization The process through which we specify precisely what we will mean when we use particular terms specific and unambiguous Conceptual Explication Variables specific concepts or theoretical constructs you are investigating Attributes O O Concepts that make up a broader concept are called attributes descriptive characteristics or qualities of variables manwoman vs gender Operationalization 99 O 9 Translation of variables concepts of interest into observable term The final specification of how we would recognize the different attributes of a given variable in the real world Philosophy and Theory in Science and Research Paradigm a fundamental model or scheme that organizes our view of something Informs our approach to doing research I Positivism Emphasizes objectivity precision and generalizability in research I lnterpretivism lndepth subjective understanding of people s lives Quantitative Qualitative Positivist Constructivist Objective Subjective Generaiizabie Speci c Numbers Wordstexts iDeductive method Inductive method Values and Ethics in Research SW510 Notes 3 Scientific Methods in Social Work What is Ethics Standards of conduct to ensure moral behavior to determine what is proper right and improper wrong behavior Historical Contexts German Experimentation Nazi Human experiments 16 physicians Nuremberg War Crime Trials Crimes against humanity Nuremberg Code 1949 Ethics in medical research voluntary and informed consent The Belmont Report 1979 The National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research mechanism for reviewing proposed research mechanism for protecting the institution and researcher IRB BEFORE any activity Social and Behavioral Sciences Institutional Review Board The Ethical Principles of the Belmont Report Respect for persons The right to privacy Voluntariness informed consent Beneficence Do no harm Maximize possible benefits Justice Research benefits Research burdens Ethical Issues Privacy Two techniques Confidentiality amp anonymity Humphrey quotTearoom Trade Study 1970 Impersonal sexual acts in public restrooms Disguise as quotwatchqueenquot and a public health interviewer for a quotsocial health survey a year later 54 menmarried 14 homosexual Ethical Issues Confidentiality and Anonymity Confidentiality How do we guarantee privacy particular information or responses cannot be publicly linked to any specific individual Exemption cases courts can subpoena records child abuse immediate risk of harm to self or others Anonymity Nobody can link individuals39 identities to their responses or behaviors Research consideration Free to participate Free to decline Free to decline in the middle of research process Really voluntary Ethical Issues Deceptions Need to justify Weighing costs and benefits Epstein 1988 bogus articles in social workjournals Telling potential participants of aH aspects consequences of research that might influence a decision to participate Written consent Effects on the quality of research Consent Form Introductory paragraph Purpose of the project research methods andor procedures to be used Duration of the project Potential benefits of participation if any Potential risks or hazards of participation if any Alternate methods of treatment or therapy if any Cost andor payment to the patient for participation Compensation and medical care in the event of injury due to this study Right to withdraw at any time without effecting treatment Contact persons and Telephone Numbers for questions or problems regarding the study Include the quotPatient Representative at 8787859 to discuss questions regarding patient rights Statement of agreement to participate Risk of physicalpsychological harm should be weighed against benefit to society Regardless of voluntary Debriefing relieving any stress Analysis and reporting stages ex Labeling Milgram quotObedience to Authority study 1963 To understand people s willingness to harm others quotfollowing the order Psychological harm to the participants Withholding treatment Control group conditions Not giving treatment or giving treatment yettobeproved Alternatives treatment group another known tx or delayed tx natural condition Ethical Issues Protecting Vulnerable Populations Justice Children severely mentally ill persons and mentallychallenged persons SW research the disadvantaged Minority Poor prisoners college students Feeling pressure to participate Research Process Ethical issues scientific misconduct Focus the consumer of the research No misleading results Fraud deliberate falsification of data Plagiarism Ethical Issues Analysis and Report 1 Report results thoroughly both positive and negative findings 2 Note limitations of study 3 Be upfront about your own biases NASW Code of Ethics Research should cause no physical or mental harm to subjects Subjects must give their informed consent to participate in the research if capable If not then assent must be obtained from an appropriate proxy Participation in research should be voluntary and participants should be informed of their right to withdraw at any time without penalty Researchers should fully disclose their identity Social workers doing research should avoid conflicts of interest and reveal possible conflicts with potential participants Avoid conflicts of interest and reveal possible conflicts with potential participants Anonymity or confidentiality must be maintained for individual research participants Findings should be reported accurately errors later found should be corrected The benefits of a research project should outweigh any foreseeable risks Social workers should take steps to ensure that participants have access to appropriate supportive services Ethical Questions Deadly Deception Was the study justified in the beginning What kinds of ethical violations are included in this research Could this study have been conducted in a more ethical manner If so how Should this study have been conducted at all Should researchers use the data generated from this study Program Evaluation SW510 Scientific methods in social work 4 Program Evaluation Program An organized collection of activities designed to reach certain objectives Royse etaL2001p5l Evaluation as a means to an end Program evaluation is the use of social research procedures to systematically investigate the effectiveness of social intervention programs Rossi et al 1999 p4 A management tool that you use to make and to help others make better decisions about social and human service programs Royse et al 2001 p 11 Program Evaluation Evaluation vs AppliedBasic Research Interchangeable Concepts Types of Evaluations Needs Assessment Can be identified Social indicators eg poverty rate Interviews with local experts Survey of population in needservice providers Structured group interviews Resource inventory Process evaluation Designed to answer questions about program operations implementation and service delivery Qualitative methods can be used Examples Program monitoring quality assurance total quality management Outcome Evaluation Deals with the core questions To assess whether or not programs accomplished goals Examples Were families reunified Are clients living independently Did the training increase skills of workers OutcomeImpact Evaluation To assess longterm change Examples Are foster families better able to cope with stress Are clients able to advocate for themselves Has quality of life improved for clients Formative amp Summative Evaluation Formative evaluation To establish the need for a program To assess initial resources design and implementation strategies During first 6 months to 1 year Summative Evaluation often used synonymously with outcome and impact assessment a definitive statement can be made about the program Examples Was the program successful or not Types of Evaluators Stakeholders Participants in Evaluation The Logic Model quotIf you don t know where you are going how are you going to get there Yogi Berra The Logic Model PE The purpose of the Logic Model is to show 1 2 The logical connections between need and the program The resourcesinputs you have to operate the program The activities for addressing need and The outputs outcomes and impact expected to result from program activities Example I ve Got A Headache Identifying Relevant Outcomes What or which outcomes are relevant for measurement Mission statement What is the program trying to accomplish or change Outcome dimension Behavior frequency magnitude duration Attitude Knowledge Efficiency Analysis CostBenefit analysis In dollars and cents Identify specific costs and benefits Challenges Whose perspectives do we use Are those cost worth it Costeffectiveness analysis Alternative to costbenefit analysis Costs are calculated and the benefits are listed Measurement Week 6 SW510 Scientific Methods in Social Work Two ways of defining concepts Nominal definitions concepts verbal definitions in which scientists agree that one set of words will be used to stand for another set of words Operational definitions the precise procedures or operations to be followed in measuring a concept Operationalization Process of connecting concepts to observations Goal To devise an operation that measures the concepts that we intend to measure Definitions of Measurement 0 process of moving from nominal to operational definition Process of describing abstract concepts in terms of specific indicators by the assignment of numbers or other symbols to these indicants in accordance with rules Process of making variables Qualities of Questions Exhaustive 0 should include all possible answerable responses 0 every case can be classified mutually exclusive 0 no respondent should be able to have two attributes simultaneously eg employed vs unemployed 0 every case can have only one attribute Level of Measurement helps you decide what statistical analysis is appropriate on the values that were assigned helps you decide how to interpret the data from that variable Level of Measurement Nominal Measurement the values just name the attribute Ordinal Measurement when attributes can be rankordered 0 Indicates relative position of cases Level of Measurement Interval Measurement distance between attributes has meaning no absolute zero that is meaningful The Hierarchy of Levels Single or composite indicators Composite Indicator indexes 339 More precise O 9 Better reliability 0 9 Better validity O 9 Increased level of measurement Single Indicator 339 Cheap and easy to administrate Measurement Caution for using a scale Assumption that each component question is INDEED measuring the same concept Combining responses to specific questions can be obscure differences in meaning among questions Sometimes particular questions are counted or weighted more than others Measurement Error True Score Theory Systematic amp Random Error Random Error any factors that affect measurement of the variable across the sample addsvanab hy not affect average performance Reducing Measurement Error Pilot test your instruments 9feedback Train your interviewers or observers unobtrusive observationmeasurement Doublecheck your data Triangulate across several measures Measurement Quality Reliability Measurement yields consistent results Prerequisite for validity Indicators of reliability TestRetest reliability Stability over time Internal consistency Splithalf reliability Parallel forms Alternate forms reliability SAT ACT Interrater reliability v ntrarater reliability Measurement Quality Validity The extent to which an empirical measure adequately reflects the real meaning of the concept under consideration A measure is valid Closely related to other valid measures of a concept Not related to measures of unrelated concepts Closely related to knownsupposed correlates of the concept Measurement Quality Validity Four approaches Face validity Content validity Criterion validity Predictive validity Concurrent validity Construct validity Convergent validity Divergent validity Face validity amp Content validity Face validity valid quoton its face a judgment call subjective the weakest form of measurement validity Criterion Validity Concurrent validity check the measure against a relevant criterion the standard is empirical The instrument is used at the same time Construct Validity construct validity is the most encompassing and comprehensive standard Others can be considered a subcase of construct validity Convergent validity measures of constructs that are related to each other should be strongly correlated Discriminant validity measures of different constructs should not correlate highly with each other Advantages I l SW510 Scientific Methods in Social Work Week 7 Survey and Instrument Survey Research A survey is a data collection technique in which information is collected by having individuals respond to questions or statements Characteristics I Data is collected from large samples of people I Respondents presented with questions I Measures are what they say not what they do attitudes and orientation Survey Research Purposes of research Survey Design Options l Crosssectional Longitudinal attraction Versatility Efficiency I generalizability Methods l Questionnaire I Interview schedule Interview Surveys Individual responds to questions asked by an interviewer who then records the answers Types Telephone Facetoface Focus groups Computer assisted motivate respondentsto give more accurate info InterVIewer can explain questions Offers moEecontrol More fleXI Illty gal observational info nonverbal responses etc ro e Telephone Interview Interviewers question respondents over the phone and then record respondents answers I Random digit dialing RDD l Interactive voice response IVR l Computer assisted telephone interviews CATIs I Pros and Cons I Electronic online Surveys Pros availability of Internet os Can be lengthy Can explain terms Attractive layouts ata processmeg low errors AaBId resc ons n nymi Cons Response rates Sampling bias Compatibility formatting across computers Interviewer comportment Rapport and control Social de5irability bias Av0id excesswe soaalizmg l Maintain a neutral and accepting attitude I Be responswe to the informant l Treat participants with respect I E31Eathy yet professional approach 0 Res on concer ns I Do not we clinical o inions Critical clinica issues repor Importance of training and monitoring Cultural Competence and the Interview Relationship Interview is a social relationship Social desirability effect impacted by gender race culture Social distance between interviewer and respondent Crossrace interviewing studies have shown large effects but not always Gender impacts on interviews Matching for race ethnicity and gender is common Language of interview has impact Things to consider Cost time and geographic restriction Sensitivity of research topic confidentiality amp anonymity Complexity of questionnaire Respondents ability to read and understand the questionnaire Possibility of interviewer bias Response rates Strengths amp Weakness of Survey Research Useful in describing the characteristics of a large population Make large samples feasible Flexible many questions can be asked on a given topic many variables Flexibility in analysis III I I3 III Same questions asked of all respondents Secondary data analysis Data gathered retrieved from I Publically available data archives I Another researcher I From a researchers previous projects Data is used to answer a research question other than initial intent of research collection Data quality is always a concern Researcher must tailor hypothesis to what is available in the data set Cheaper and faster Disadvantage validity issue Types of Question Openended No prepared response choices Interview scheduleinterview guide Exploratory research Advantages richness of information no category willingness to answer Disadvantages timeconsuming data processing lengthyillegible responses coding Types of question Closedended Predetermined response set amp is limited Two structural requirements deciphering Important that responses are mutually exclusive and exhaustive or have check all that apply option Questions may be skipped if respondents do not find an appropriate response Advantage Easy data processin uniformity reduces ambiguity Disadvantage lack of Indepth In ormatlon Basics of questionnaireinterview schedule Does it get the information necessary to test the hypothesis It measures knowledge attitudes feelings or symptoms attributes NOT behaviors The avoidance of measurement error I Systematic error I Random error Errors Errors of observation poor measurement of cases surveyed Errors of nonobservation omission of cases that should be surveyed Other errors I Question wording l Respondentcharacteristics l Presentation of questions I Interviewer Result Writing clear questions Lose validity if meaning is not conveyed to respondents Must be understood by diverse populations Consistency in survey questions not varied by individual Be aware that questions are asked in social cultural and economic context Important that same idea can be understood by two very different individuals Guidelines for Asking Questions Avoid confusing phrase job concerns vs Avoid vague words time place Avoid jargons Avoid doublebarreled Questions Avoid negatIve Itemsdouble negatIve Wording is Important Roper Poll Does it seem possible or does it seem impossible to you that the Nazi extermination of the Jews never happenedquot 34 uncertain Gallop Poll in your opinion did the Holocaust definitely happen probably happen probably not happen or definitely not happenquot 79 definitely happened 17 probably happened Guidelines for Asking Questions Respondents must be willing to answer Respondents should be competent to answer Questions should be relevant topics they care about Short items are best I Read quickly and answer without difficulty r l Survey the general populations Write at a 3 grade level llEikert Sctaling d re uen use In surve ues 0 hair A sceqriesof ositio s atexignts tf39oli owec w 4 or 5 response alternat ve Lecye o a ree ent ents In Icate t eIr c orce from the unambiguously ordered es on Jesslgciiyslearcegteergtceldewith statements level of agreement Semantic Differential Posing two opposite positions Generally 7 points Filter or Contingency Questions Certain questions will only relevant to some respondents Matrix Questions Series of questions that have a common theme and same response choices Shortens questionnaire Emphasizes common theme BE AWARE I Be very clear in instructions so respondents do not think they need to complete the entire matrix l Response set people might not read the question and just circle the same thing down the entire survey l A Checklist of Considerations I The Opening Questions I easy to answer nonsensitive nonthreatening I never start mail survey with an openended question I get the respondent rolling l Selfadministered interesting one first I Interview building rapport relationship I Sensitive Questions only after trust is developed I put demographics at end unless needed to screen I for historical demographics follow chronological order I A Checklist of Considerations l Include clear instructions and introductory comments when appropriate I At the beginning and anytime format changes I Example V or rank order I Provide boxes for each question brackets or parentheses circles consent form vs introductory paragraph Here are some things the government might do for the economy Circle one number for each action to show whether you are in favor of itquot A Checklist of Considerations Ask about one topic at a time Use a transition when switching topics Decisions About Placement l Interesting one first selfadministered I Is the answer influenced by prior questions Explain skip patterns clearly filter or contingency questions make a flowchart General rules Keep questionnaire as short as possible Visually attractive spread out and uncluttered 5 pages or shorter What information do you want to get Why are you asking this question What concepts are you measuring Guidelines for Addressing Cultural Sensitivity Reliability and validity with the culture Use bilingual interviewers Modify existing instruments if none available for culture studying Language difficulties eg back translation concepts Pretest Knowledgeable informants to assess problems Immerse yourself in the culture before selecting constructing administering SW510 Scientific Methods in Social Work Week 9 Sampling Lecture overview Sampling Terminology Population vs Sample Parameter vs Statistic Sampling frame Why do we study with sample Probability Sampling Designs NonProbability Sampling Designs How to determine sample size What is sampling Selecting a group of peopleobjects sample from those who we want to study population in order to find out population characteristics parameter based on sample characteristics statistic Sampling Why do we use a sample Time and Money Population is changing Why study sampling Are the findings true for only those who provided the information or is it about the population Sampling Terminology Population The entire set of individuals or other entities to which studyfindings are to be generalized Elements The individual members of the population whose characteristics are measured Sample A subset of the population that is used to study the whole Sampling Frame a listing of all the elements in a population Census a study of the entire population Parameter Summary description of a given variable in a population Statistic Summary description of a variable in a sample Generalizability Representativeness a criteria for the quality of sampling A sample will represent its population if the sample s aggregate characteristics closely approximate those same aggregate characteristics in the population Sampling error estimate of any difference between the characteristics of a sample and the characteristics of a population from which it was drawn sample size amp homogeneity The larger the sampling error the less representative the sample Sampling istributions amp Sampling Error Sampling distributions a hypothetical distribution of a statistic across all the random samples from a population The mean of the sampling distributionthe true mean of the population used to estimate sampling error Sampling error Average distance of sample means from the true population mean ProbabilitySamplingPS PS Simple Random Sampling Identifies cases strictly on the basis of chance Random CI Each element has an equal chance of selection CI independent of any other event in the selection process Procedure CI choose sampling frame CI Assign a number to every unit in sampling frame CI decide sample size that you want or need n CI use Table of Random Numbersquot PS Systematic Random Sampling Assumes that the population is randomly ordered Advantages efficient Weakness IF the list of elements a pattern Procedure Assign a number to every unit in sampling frame 1 to N decide sample size that you want or need n Nnk the interval size First randomly select a number from 1 to k Then take every kth unit PS Stratified Random Sampling Modification of simplesystematic random sampling sampling strata All elements in the sampling frame are distinguished according to their value on some relevant characteristics The size of each stratum in the population must be known Reduces sampling error proportionate disproportionate PS Cluster Random Sampling The final units to be included in the sample are obtained by first sampling larger units clusters in which smaller units are contained Useful when sampling frame is not available Step1 Divide population into clusters Step2 Draw a random sample of clusters Step3 Elements within randomly selected clusters are randomly selected Efficient but prone to sampling error Nonprobability Sampling Designs NPS Pros and Cons Types No known probability difficultimpossible to detect misrepresentation Does NOT yield representativeness cheaper and faster Availabilityconvenience Purposive Quota Snowball NPSAvailabilitySampling Accidental haphazard or convenience Sampling available or accessible clients weakness we have no evidence for representativeness voluntariness NPS Purposive Sampling uses judgment and prior knowledge for the sample best serve the purpose of the study Selecting informants Knowledge of areaarenasituation under study Willingness to talk Representativeness of range of points of view an extreme group Exploratory study Completeness Saturation NPS Quota Sampling Preset number of elements based on characteristics in a population to ensure that the sample represents those characteristics in proportion to their prevalence in the population ex Womenmen Once a quota is reached no more elements from that category are put in sample Similar to stratified Interviewers select the sample interview people until they have met all of the quotas on each variable NPS Snowball Sampling One person recommends another who recommends another who recommends another etc Useful hardtoreach populations Subculture Sensitive topics How to determine sample size homogenous populations vs diverse populations Stratified random samples vs simple random samples kDOOlCDU39 Complexity of the analysis Strengths of relationships among variables statistical power analysis NPS What to consider State your hypothesis What populations or sources of data will best describe the relationships in your hypothesis How do these populations or sources define the relationship beyond your sample Meaning how are you going to generalize your findings Are there differences between your proposed sample and the characteristics of the population you are studying Describe any resource limitations that you may have What is the potential population size What sampling frame will you use How will you select your sample What is a prudent sample size 10 Are there any class race or gender biases represented by your sample 11 How will you address these biases in your study 12 Describe your final sample Causal Inference and Group Designs SW 510 Scientific Methods in Social Work Outline week 10 Causality Longitudinal and crosssectional design Unit of analysis Threats to external and internal validity Major group designs Establishing Causality Under what circumstances can we infer that a change is caused by the IV o Is our program intervention effective 0 Does our program cause desirable results improvement 0 Does X independent variable cause Y dependent variable not something else 0 What are the most plausible rival explanations How do we rule them out Criteria for inferring causality Time order the cause precedes the effect in time Association The DV and IV should be empirically correlated with each other rd Nonspuriousness relationship is NOT due to a variation on a 3 variable The Time Dimension Cross Sectional Studies vs Longitudinal Studies Crosssectional study A snapshot at one point in time Statistical control attempt to rule out the plausibility of rival hypotheses by controlling for alternative variables Multivariate analyses methods Unit of analysis amp errors in reasoning Units of Analysis the level of social life on which the research question is focused such as individuals groups towns or nations The Ecological Fallacy and Reductionism Ecological Fallacy o a group level premise is used to draw conclusions about individuals 0 drawing conclusions about individuallevel processes from group level data Reductionist Fallacy 0 Making inferences about group processes from individual data reduces individual behaviors to being entirely caused by individual attitudes and preferences The confidence we have that the results of a study accurately depict whether one variable IV is a cause of another DV An experiment s ability to yield valid conclusions about causal effects 1 Existence of comparison groups or control 2 The comparability of experimental and comparison groups Preexperimental designs Notation X stimulus intervention 0 observation oneshot study onegroup pretestposttest design posttestonly with nonequivalent groups SingleGroup Threats to Internal Validity History threat 0 any other event that occurs between pretest and posttest Maturation 0 Normal growth between pretest and posttest Regression to the mean 0 group is a nonrandom subgroup of population Testing 0 the effect on the posttest of taking the pretest Instrumentation 0 any change in the test from pretest and posttest Multiplegroup threats to internal validity Contamination Awareness of existence of other groups Compensatory rivalry 339 controls compete to keep up with treatment group Resentful demoralization 339 controls quotgive upquot or get discouraged Diffusion of treatment 339 controls might learn about the treatment from treated people compensatory equalization Multiplegroup threats to internal validity Mortality attribution Threat nonrandom dropout between pretest and posttest EVEN WITH random assignment groups may become different over time because of MORTALITY differential attrition More time experiment takes the higher the chance of differential attrition Experimental Designs To establish causality controlling for threats to internal validity Manipulate and isolate IV Randomization 339 Random assignment Two groups must be comparable 339 Introduction of intervention to one group 339 Comparing the amount of change in DV Matching comparability 339 Matching characteristics 339 Random assignment Multiplegroup threats to internal validity Selection bias Two groups different Related to methods used to assign the subjects to group the groups are comparable Issues of how you assign units eg persons to the groups or select them into groups group improved much more or less than the other group because of initial group differences NEED random assignment Classic experimental design ED Posttestonly Control group R X 0 R 0 Random assignment probabilistic equivalence between groups Introduce the independent variable experimental Compare posttest scores between the two groups differences between groups 9 a treatment effect Remove testing threat usually test this with a ttest or oneway ANOVA ED Solomon four group design ED alternative treatment design with pretest True Experimental Designs in agency settings The experimental program cannot change during the course of treatment Implementation dependent on staff with different skill levels More controlled less generalizable Threats to professional judgment People are suspicious of quotnew treatment quotIf it is so good why can t my clients get it now Quasiexperimental Design Experimental design feasibility comparable to the treatment group in critical ways but are not randomly assigned No randomized assignment to control and experimental groups Weak internal validity Comparison group not control group How to increase validity to QE QE Timeseries design Simple interrupted timeseries design Similar to classic experimental design No control group or comparison group Time as a controlling variable Multiple timeseries design no randomization no control group When true experiments are inappropriate SW 510 Scientific Methods in Social Work Week 11 Single Subject Designs Single Subject designs Time series designs used to evaluate practice with a single case individual group family or community N1 Singlesystem or singlecase designs Pavlov s dog The Logic of Single Subject Designs Time as controlling factor Baseline Repeated measures during baseline A controlling phase and intervention phases B experimental phase Unlikely shifts in data trends or patterns coinciding with changes in phases Problems should be operationalizable Primary limitation external validity 1 Measurement Operational definition of problem and goals Unobtrusive 0 Direct observation 0 Existing records Obtrusive o Paperandpencil measurement scales 0 Se freport o Triangulation 0 Measure of indicators 23 indicators 0 Use multiple observers if possible Basic components What to measure Outcome behavior Not negative but positive indicator Frequent events Multiple indicators 23 indicators Who should measure Practitioner Clients selfmonitoring Relevant others Data quantification procedures Frequency Duration Severitymagnitude 2 Establish Baseline enough repeated measures 510 points at least a week a stable trend internal validity Retrospective baseline reconstructed easy to recall immediate past Baseline Stability Stable baseline 0 Increasing trend Decreasing trend Cyclical Unstable 3 Introduce treatmentintervention Single treatment or a combination of coherent treatments required for evaluation Continue to monitor singlesystem s behavior AB Design Basis of other models Single baseline period single intervention period B Design ABA Reversal and Withdrawal Withdrawal design treatment sto s entirely during an A phase Reversal desi n treatment trans erred to another behaVIor durIng A phase ABA and ABA options Demonstrate treatment effect Multiple Baseline Monitor two or more situations simultaneously Stagger intervention Fairly strong on internal validity Carryover effect Principle of unlikely successive coincidences Visual Analysis Graph data Clinical significance Statistical significance Threats to lnternal Validity History Maturation Testing Instrumentation Statistical Regression Dropoutmortality Generalizability Direct replication Systematic replication Replication Complicating Factors Incomplete data baseline 0 Not long enough A or B 0 Ethical ssue elay of tx O Retrospective baseline Generalization effect 0 Apply one skill to other areas 0 One Intervention result affects other areas Carryover effects 0 carrying over to next phases ex ABC Order effects o The order of Interventions Practicalfeasibility constraints Client s crisis baseline Heavy caseload resources Lack of support from other practitioner or agency level Clients resent selfmonitoring Efficiency accountability compassion O Compassionate conscientious and professional practitioner 0 Commitment to welfare of clients 0 Can we afford to use untested interventions not knowing whether we are helping or harming clients or potentially wasting scarce resources SWSlO Scientific Methods in Social Work Week 12 Qualitative Research Qualitative Research Data cannot be easily reduced to numbers Researcher39s conviction to capture the actuality Researcher s underlying strategies What we know is a mixture of facts and point of view Nature of the problem exploratory Clarify and illustrate quantitative findings Build new research instruments or a theory IVIethods participant observation indepth interviewing focus group study ethnography case studies Qualitative and Quantitative study Similarities The goal of research to understand the world better Research consists of logic and observation data Differences Qualitative research induche holistic subjectiveinsider centered process oriented anthropological worldview Quantitative research hypothetical deductive pa rticularistic objectiveoutsider centered outcome oriented natural science worldview Qualitative and Quantitative study Qualitative research goal understand actor39s view dynamic reality assumed discovery oriented relative lack of control exploratory Quantitative research goal find facts amp causes static reality assumed relative constancy in life verification oriented attempt to control variables Confirmatory explanatory Research designs Participant Observation Indepth Intensive interviewing Focus groups Ethnography naturalistic observation Case study single unit examination information about an individual case is systematically organized and presented in the form of a narrative summary Qualitative Methods The nature of the data Exploratory research questions inductive reasoning Unstudied processes and unanticipated phenomena An orientation to social context to the interconnections between social phenomena Human subjectivity meanings that participants attach to events and that people give to their lives Reflexive research design Sensitivity to the subjective role of the researcher Sampling in Qualitative Methods Strategies Nonprobability strategies purposive sampling snowballsampling quota sampling Theoretical strategies 39 systematic approach particular processes that seem to be important and select instances to allow comparisons or checks with which perceptions can be tested adding members to sample to include additional critical typical or deviant cases Qualitative Methods Fieldobservational research process Establish specific goals of research Review the relevant literature Decide upon role of researcher Make use of informants Gain entry into the group for study Sample your population Develop rapport with persons Begin observations EDP NP Pquot39gtP Record observations 10 Data analyses Role of researcher in participant observation Complete participant Participant observation Complete Observer covert participation Variations in between start as outsider and move to membership Or change to outsider role at end to verify hypothesis generated as a participant Data collection methods Tape recording Taking notes Primary means of recording participant observation data 0 Jottings brief notes taken in the field 0 Field notes written after observations based on jottings 0 Daily log daily record of activities sentiment and perceptions Client logs Advantages Interaction between data collection and data analysis flexibility Disadvantages selective observations Qualitative Methods Interviewing Three forms of openended interviewing Informal conversational interviews spontaneous using p 39PII probing questions how IS that or In what way Interview guide approach Standardized openended questions Qualitative methods Intensive interviewing relies on openended questions the specific content and order of questions can vary from one interviewee to another Goal to develop a comprehensive picture of the interviewee s background attitudes and actions in her or his own terms follows a preplanned outline of topics simple questions that gather background information while building rapport followed by a few general grand tour questions ndepth interviewing Selecting respondents 0 Random selection Knowledgeable willing to talk Represent a range of perspectives 0 saturation point at a point when new interviews yield little additional information 0 Theoretical sampling systematic approach select new settings or individuals that permit comparisons to check their perceptions Qualitative methods Focus groups gt Goal Learn about beliefs attitudes and opinions about socialpsychologicalcultural characteristics F Format 0 Group discussion of unrelated individuals710 b Facilitators ask questions and guide the discussion gt Sampling 0 have the time to participate 0 have some knowledge pertinent to the focus group topic 0 share key characteristics with the target population Capitalizing on synergistic group effect Strengths amp Weaknesses Depth of understanding gt External validity 0 Analytical generalization connecting case study findings to a particular theory gt Subjectivity biases and points of view understanding world view of respondents Attempts to avoid pre judgments gt Dependent upon researcher39s personal tt 39b t d k39ll FlexibilityampOpenness a n u es an S I S gt Participation can change the social InexpenSIve staffingamp Situation equipment vs paper amp pencH Threats to Trustworthiness 39 ReaCtiVitV gt Prolonged engagement Researcher biases gt Triangulation corroboration Respondent biases between two or more social desirability SOUFCES gt Peer debriefing and support gt Negative case analysis gt Leaving a paper trail for auditing Similarities to SW practice Begin where the client is Focus on perceptions of the informant Attempt to understand environmental context Try to avoid imposing preconceived ideas or theories Ethical Issues Voluntary participation Subject wellbeing Identity disclosure Confidentiality Appropriate boundaries Constructing Qualitative Measures Depth of meaning from the respondent s perspective IVIore flexible interview schedules Mostly interviews The format of qualitative interviews varies Completely unstructured informal conversational interviews Semi structured interviews Interview guides outlining topics Highly structured standardized interviews SW51O Scientific Methods in Social Work Week 1 Introduction Overview 1 Introduction Syllabus 391 What is Social Work Research 391 Evidencebased practice 1 Purpose of research Researchamp Other ways of knowing Tradition shared meaning and understanding that is often considered obvious Authority knowledge accepted based on the status or power of the messenger Common sense reasoning or commonly held beliefs Popular media television Internet newspapers and other popular sources Expenence Intuition Flaws in Unscientific Sources Overgeneralization the assumption that a few similar events are evidence of a general pattern Selective observation the tendency to pay attention to future events and situations that correspond to or confirm a pattern perceived to be true Illogical reasoning Premature jump to conclusions arguing on the basis of invalid assumptions Ego involvement occurs when personal involvement or investment in a particular result clouds objectivity Premature closure of inquiry Inaccurate observation Ex post facto hypothesizing Social Workers amp Research Social work researchers can produce knowledge that practitioners find useful To provide practical knowledge that social workers need to solve problems Social work practitioners can use research techniques that help guide their practice To be good consumer of the research literature Why is Social Work Research Important To solve specific practical questions Inform good practice Accountability and compassion To serve our clients better The amount and types of services you promised Identify problems and unmet needs To create and disseminate knowledge for use by other professionals It is part of our CODE OF ETHICS NASW 502 research and evaluation Research The active diligent and systematic process of inquiry aimed at greater understanding of events behaviors or social phenomena Basic or pure research advance knowledge with little concern for immediate practical benefits Applied Research practical outcome is the focus assume some group or society will gain specific benefits Social work research Social research is the systematic examination or reexamination of empirical data collected by someone first hand concerning the social or psychological forces operating in a situation 39lllj quotl i i l l 7 Social work research Social research is the systematic examination or reexamination of empirical data collected by someone first hand concerning the social or psychological forces operating in a situation 39lllj quotl i i l l 7 Scientific methods to errors Overgeneralization Systematic sampling ogica reasoning Specific criteria for causality Selective and inaccurate observation Systematic observation and measurement Ego involvement Peer evaluation Remember In science everything is open to question Knowledge is always subject to change Science is also subject to error Social Work Research in Practice Purposes Exploration Description Explanation Evaluation Purposes of Research Exploratory Research Qualitative research methods commonly used Investigation of social phenomena without expectation What is it like to be homeless How do people get along in the situation What concerns them Purposes of Research Descriptive Research Survey research methods commonly used Gathering of Facts Who are the homeless What are their needs How many people are homeless Purposes of Research Explanatory Research Experimental research methods commonly used Seeks to identify causes and effects Why do people become homeless Depends on our ability to Show that the two events are related Demonstrate a time order between two events Rule out alternative explanations Purposes of Research Evaluation Research also referred to as practice evaluation or program evaluation What services help the homeless Involves searching for practical knowledge for consideration of Implementation of social policies Effects of social policies Impact of social service programs EvidenceBased Practice EBP Evidencebased practice is a practice model based primarily on the scientific method and scientific evidence This model Encourages the integration of scientific evidence practice expertise and client circumstances Evaluates outcomes Is applicable for intervention assessment policy and community level practice Client centered critical thinking lifelong learning understanding research and design EBP amp social worker EBP suggests that social workers should integrate the best current research evidence knowledge derived from research to achieve a particular outcome Evidencebased practice Steps Question formulation Searching for evidence Critically appraising the studies you find determining which evidencebased intervention is most appropriate for your particular clients Applying the evidencebased intervention Evaluation and feedback The Cycle of EBP Client state Appropriate for aiggges jearlt 0 this client circumstances Actuarial risk assessment Current Other valid Pest assessment EV39dence measures Client Preferences Client s and actions preferred course or at least willing Effective services to try Barriers eg cultural conflict Quiz Identify purposes of research in the followings 1 I have proposed a project in which I will interview several immigrants from other countries to find out why they chose to move to the United States 2 A recent survey found that only 23 of social workers use the internet to provide services compared to 48 of professional counselors 2 Evidencebased practice The social worker at an addiction treatment program is trying to work with court mandated clients What should she consider
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'