Research Notes PDF

Research NOTES Week 1 Evidence-Based Practice: ○ Practitioners adopt a lifelong learning - involves posing specific questions of direct practical importance to clients searching objectively and efficiently for the current best evidence relative each quest...

Research NOTES Week 1 Evidence-Based Practice: ○ Practitioners adopt a lifelong learning - involves posing specific questions of direct practical importance to clients searching objectively and efficiently for the current best evidence relative each question and taking appropriate action guided ○ Practitioners integrate best research evidence available in their practice with clients attributes,values,preferences ○ Do no harm ○ Informed consent Steps in evidence-based practice: ○ 1. Convert information needed-prevention ,assessment ,treatment risk into PICO question ○ 2. Track down best current evidence ○ 3, critically appraise evidence ○ 4. Integrate critical appraisal with practice experience,clients preferences, values and circumstances ○ 5. Evaluate effectiveness and efficiency in exercising 1-4 seek ways to improve them ○ 6, teach others to follow same process Evidence-based practice and macro social work ○ EBP rooted in clinically oriented domaIn but many aspects are adaptable for community,organizational ,administrative and policy social work practoce ○ Macro perspective: EBP =decision making process involving policy makers,purchases,managers key stakeholders Benefits of using EBP in social work ○ Enhances quality ○ Empower client + practitioners ○ Develops thoughtful and ski;lled practitioners ○ Adds social work contributions to scientific literature ○ Builds bridges across professions ○ Honours ethical obligations and profession standards Concerns about EBP in social work ○ Ebp cookbook approach doesnt consider complexity and difference ○ May undermines traditional professional practice ○ Being reductionistic,mechanistic ○ Rationing disguised as science ○ Week 2 PICO = population,intervention comparison outcome =Quantitative  ○ Population and their problem How would i describe a group of clients of similar ○ Intervention: what might you do? Apply a treatment ,act to prevent a problem ○ C: Alternate course of Action: what is main alternative other than in box ○ O: what you want to accomplish Outcome of treatment or prevention accurate description of client ,valid measure ○ Example of PICO question: For low-income families with school age children do asset/wealth building interventions compared to no intervention increase the likelihood of sending their child wren to college in the future PICo = population , interest context = Qualitative ○ P: client type, population and or problem ○ I:What is the phenomenon of interest ○ Co: What is context and the outcome we want to understand ○ Example of PICo question: if caregivers who provide care to their older relatives are interviewed how would they describe their experiences of providing care Characteristics of a PICO question: ○ Client oriented: begin matter of important to client ○ Practical importance: pertain to Social work practice ○ Guides in the search for evidence : framed specifically Week 3 What is qualitative research: ○ Understanding of underlying reasons opinions and motivations of people ○ Describe an experience or phenomenon ○ Situated activities that locates the observer in the world ○ Consists of set of interpretive , material practices that make world visibe Assumptions of qualitative research ○ Interpretive /constructivist paradigm Reality is interpreted and or constructed Reality subjectively discerned Reality is situated (historically ,gender ses) Qualitative studies provide different quantitative ○ Focuses on individual subjective experiences rather than predetermined categories ○ Seeks to understand perceptions values,attitudes and beliefs ○ Allows for the development new theory or addition to existing theory Uses of qualitative research  ○ Explore topic about which is little is known ○ Pursue a topic sensitivity andd emotional depth ○ Capture the lived experience ○ Seek to merge activism with research Qualitative research and EBP ○ Can also compliment quantitative research by providing info for Generating hypotheses Developing and validating instruments Providing context for evaluating evidence-based interventions Improving intervention designs ○ Qualitative research and EBP in practice Used for: Understanding and explaining the potential trajectory of a healthcare condition or social problem Building empathy and therapeutic alliance with clients Improving an understanding Qualitative methodologies ○ Many research designs of strategies used to collect data ○ Should specify questions answered ○ Consistent with research question, intent of research Qualitative research family ○ biography(narrative) ○ Phenomenology Lived experiences within the world ○ Ethnography Collecting data in social and behavioural sciences Through observations + interviews Ex: researcher living and working with community in an isolated village or tribe ○ Case studies explores in depth a program, event, activity, process, or one or more individual ○ Grounded theory Explanations are grounded in participants own interpretations or explanations ○ ○ Participatory action research  researchers and participants working together to understand a problematic situation and change it for the better Sampling ○ Random sampling not considered ideal in qualitative research ○ Researchers use non probability sampling can include Purposive intentionally selecting participants based on their characteristics, knowledge, experiences, or some other criteria. Snowball provides a natural way to collect data through the connections of the agents in the social network. Ex: collecting info from facebook or twitter Quota sampling method for selecting survey participants that is a non-probabilistic version of stratified sampling ex: if there are 100 men and 100 women in the population and a sample of 20 are to be drawn, 10 men and 10 women may be interviewed. Maximum variation sampling contains cases that are purposefully as different from each other as possible. ○ Number of participants may not be predetermined Qualitative process of Data Analysis ○ Collect the data Observation notes, taped interviews fieldnotes ○ Organize the data Transcribe the taped interviews fieldnote memo ○ Read through raw dara Obtain general sense of the material ○ Code the data Develop open coding ,axial coding and matrix Coding process ○ Step 1 identify concepts ○ Step 2: group into categories ○ Step 3: statements of relationship ○ Step 4th : substantive level theory Data collection methods ○ Focus groups: Group discussions of 6-10 participants Members interact with each other to generate depth and richness of ideas  ○ Artifact examination Anything physical evidence that may help provide information to understand phenomenon Written documents: personal writings, informal correspondence ,electrronic messages Records: birth certificates, annual performance reviews formal letters Objects: gifts photo album Data collection sample size: ○ Stop data collection when Saturation is reached ○ No new information about a category emerges during coding ○ Properties and dimensions are well developed Theoretical saturation- no new relationships Transcription of data: ○ Transcribe every word ○ Tone of voice: Loudness= bold or capital letters ○ Voice pitch (optional) : high/low arrows ○ Pause: ○ Non-verbal: laugh smilee ○ Data analysis : use of diagrams ○ Begin to use relationships between concepts start to emerge ○ Start with simple drawings that connect specific ideas ○ Diagrams increase in complexity as larger picture emerges ○ Core concept Reporting the findings ○ Using “ thick description ) ○ Using examples from text :(quoting the participants to support conclusions Methodological criteria/rigor for qualitative research design ○ Is the type of qualitative study explicit (phemonlogical ○ Are the methodological component clearly described (sample data collection,data analysis Methodological criteria for data analysis and findings ○ Are the steps taken to analyze the data described and appropriate ○ Is evidence provided to support the findings ○ Are the findings plausible do they make sense Trustworthiness in qualitative research ○ Credibility: truth value- findings believable ,accurate correct Prolonged engagement- history of involvement Persistent observation Triangulation: two or more sources to achieve comprehensive picture Peer debriefing: regular opportunity to discuss data collection Reflexive journal  ○ Transferability : is there enough info for someone to apply findings to another setting Eg: if cognitive behavioural therapy is effective in treating adolescent girls would it be helpful for treating depression in middle aged men ○ Dependability: appropriate procedures used to ensure reliable results eg: if study was repeated by different researcher would they find the same results ○ confirmability : neutrality- can the findings be verified Audit trail: allows someone to review products and processes of study Extensive record: systematic description of process of study Transcripts ,sampling decisions ○ Generalizability: do the findings apply to entire population regardless of context (uusing anti-depressants for people with depression ○ Week 5: Intervention: Activity or set of activities designs to cause a change in an individual, group or community Designed to cause change in some type of outcome or dependent variable Under control of and manipulated by researcher of study Intervention itself = independent variable ○ Outcome variable(dependent variable is assumed to depend on or be caused by intervention ○ Intervention effects or intervention outcome Causality: When one even is responsible entity for producing change in something else or outcome ○ association/correlation: independently variable(cause) and dependent variable (outcome) must be related to each other ○ Time ordering/temporal relationship: independent variable must occur earlier in time than the dependent variable ○ Absence of confounder: observed relationship between two variables cannot be explained due to influence id a third variable Confounding variable: ○ Alternative change cause to change in a particular outcome or set of outcomes ○ Occurs concurrently with the intervention and produces own effects ○ Could be the re  Internal validity: ○ Was the intervention,program or policy the most plausible cause of the observed outcome or is some alternative explanation plausible? ○ Degree to which the intervention itself caused ○ Time order: provision of intervention must precede or coincide with change in the outcome being measured ○ Correlation: changes in the outcome measure must be associated with the changes in the intervention condition ○ Ex: in a study there is a correlation between smoking and being a low income Threats to internal validity: ○ History: The possibility that other events may coincide with provision of the intervention ○ Passage of time or maturation Explained with many survivors decrease in their symptoms Maturation: how the patient feels ○ Statistical regression of the mean: If there are extreme cases ,values ,otherwise known as outliers, these values/cases tend to become less extreme overtime and closer the average regardless of intervention Ex:fighting in toronto public school decrease not because of intervention but because fighting was high for couple weeks ○ Testing effects: When testing itself enhances or alters the performance when being tested but doesnt change the outcome of intended behavior Ex: youth are tested once a week for duration of intervention ○ The youth realize there are certain answers want to see so they choose these answers instead of changing their behaviours ○ measurement /instrumentation: Changes in measures themselves or changes in the way they are used or applied overtime if there are any differences between the treatment and control group  Ex: the research team decides to change the measurement tool to measure bullying , name calling mid way through the intervention- looks sucessful but it is attributable to change in measurement/ instrumentation ○ ○ Statistical conclusion validity ○ What is the probability that the affectives or lack of that can be attributed to statistical chance Selectivity bias(selection) ○ When the treatment and control groups differ with respect to determinants of the outcome therefore differences in the intervention and control group attributable to selection bias and not the intervention Ex: toronto public schools intervention for middle school youth- parental involvement is higher for the youth who are in the intervention group than the control group Parental involvement is critical in decreasing youth violence Attrition ○ When research team fails to consider the outcome of participants who dropout of intervention (either the treatment or control/comparison group Ex: research tram concludes the TPS youth violence intervention is effective but you notice so many of participants dropped oout of intervention and team doesnt know why - dont know if intervention is successful Casual ambiguity: ○ When there is ambiguity about the direction of casual effects of intervention ( AB) EX: In the TPS Youth intervention - notice that addressing loneliness to improve anxiety symptoms but it could be by improving anxiety you improve loneliness Control group contamination: ○ When control group receives experimental treatment Ex: In TPS intervention students who are receiving intervention or in control group learn about breathing exercises to control their anger - those students mingle with those in the control group- control group students are using breathing exercises noe Randomization ○ Randomly assigning research participants by chance to receive different treatment groups or control/comparison group  ○ Important in intervention research - allows team to take into account potential; differences between treatment and control group to determine the anticipated change in outcome is because of intervention ○ Helps prevent bias -key defining feature in which differentiates experimental designs from non experimental designs Equal External validity: ○ Degree to which results of a study hold true in other settings ○ As known as generalizability ○ Degree Types of research designs: Pre-experimental: ○ One shot case study: One group/(intervention treatment only) where respondents are observed after intervention is finished but not before the intervention occurs T1/X T2/O2 ○ One group pre-test , post test study design One group (intervention treatment only) and respondents are observed before the intervention of conducted and after intervention is finished T1/O1 T2/X T3/02 Experimental designs ○ Pretest post test control group design Two groups -each respondent is randomly assigned to treatment or control group Research team observes/measures the treatment group both before and after the experimental treatment/intervention Also observes/measures the control group at the same time as treatment group but don't receive experimental intervention/ treatment Limitations: testing effects( the pretest itself influences how respondents take the post-test - creates bias ○ Post-test control group design  Randomly assigns research participants to treatment and control groups observe /measure treatment group after experimental treatment/intervention observe/measure control group at the same time as treatment group but dont give intervention treatment Benefits: good to use if pretest not available Limitations : no way to check if the groups are equal at baseline or determine the magnitude of pretest or posttest-gain Solomon four group design ○ Combines pretest-post-test control group design + post-test only control group design ○ Considered the best for experimental research because research team is able to show Pretest to post-test gains (comparing first two rows) Simultaneously examine testing effects (comparing all rows together ○ Limitations: Given thar research team needs four separate groups(two intervention two control ) to use design it is expensive Alternative treatment pretest and post-test design ○ First randomly assign participants to treatment /control groups ○ Research team would measure treatment groups before and after their respective experimental treatments/intervention ○ Research team would measure control group at the same time but not receive intervention Benefits: This team can assess the relative improvement for each intervention and demonstrate whether either treatment is better than no treatment (or control group) Quasi-experimental research designs When impossible or unethical to assign subjects to experimental and control groups in real world settings quasi experimental research designs may be used Quasi experimental designs provide credible although less than ideal alternative research is not feasible Those designs are frequently more practical than experimental designs and apply more directly to the real world Key difference between quasi-experimental(no random) and experimental research designs= randomization Have less control than experimental ○ Non-equivalent control group design Researchers administer pretest and post-test to both treatment group and non-equivalent but similar control group or comparison group  Advantages: superior to one group pretest posttest design because research team can determine if treatment and comparison group have comparable levels to the outcome variables and comparable levels to confounding variables (eg demographic variables) ○ Simple interrupted time series design No comparison group is needed - used to evaluate ones own therapeutic practice ○ Treatment group: make series of observations on treatment group both before + after intervention is administered Comparison group: series of observations on comparison group at the same time as treatment group but dont recieve intervention Design has greater internal validity than the simple ○ Intervention Fidelity defined as extent to which an intervention was delivered as conceived and planned to arrive at valid conclusion concerning its effectiveness in achieving target outcomes Degree to which program both implemented prescribed elements and did not implement non prescribed elements Degree of which the intervention is carried out as per instruction Establishes how well the treatments of the intervention were delivered Aka: treatment adherence,program integrity , program fidelity How to assess intervention fidelity ○ Before intervention: Training on intervention planning,delivery and execution ○ During or after intervention Video several randomly selected sessions with new SW Strongest -> weakest types of research 1. Systematic reviews and meta analyses 2. Multisite replications of randomized experiments 3. Randomized experiment 4. Quasi- experiments 5. Single case experiments 6. Correlational studies