research design exam 2

descriptive research

document the nature of existing phenomena (traits, behaviors, conditions)

can provide data for development of hypotheses that can be tested by future studies

surveys and secondary analysis of databases are commonly used

inquiry is

deductive and hypothesis-based where researchers select variables of interest

surveys

can test hypotheses about the nature of relationships within a population

can be descriptive - characterize knowledge, behaviors, patterns, attitudes, or demographics of individuals within a given population

questionnaires

structured survey that is usually captured via self-report

designed and standardized in consultation with research experts so that everyone is exposed to the same questions in the same way

easy to distribute

potential for misinterpretation of the questions and unknown accuracy from respondent

does a questionnaire need to go through the IRB?

it depends <3

depends on nature of the research, policies of the institution

may need approval to protect confidentiality and privacy of human subjects

what is the first step in developing a questionnaire?

identifying the research question

how should you word questions in a questionnaire?

use purposeful language (short, simple terms)

avoid bias

clarity (understand what you are asking)

avoid double-barreled questions

consider frequency and time

sensitive questions

double-barreled questions

asking two things in the same question

not using and/or

avoid this in making questionnaires

cognitive interviewing

ensuring that the population understand the survey questions

what is their interpretation of the questions?

qualitative research

explaining things, not testing a hypothesis

favors a world view that strives to represent social reality as it is found naturally

inquiry is inductive: observation --> explanation

uses data collection methods including open-ended discussion and observation

no control of variables

true or false: generalizable, statistically significant, and objective finding are the best sources of evidence for practice

false :)

objective data can still have bias

subjective is helpful for understanding personal experiences

qualitative = deeper, more detail, more understanding

still valuable information

true or false: the perspectives of the few do not represent many. large sample sizes are needed for clinically meaningful information

false !!!

fewer participants is not always a bad thing

depends on the study design

true or false: self-reported data is unreliable; qualitative data is only useful when supported by numbers

false

qualitative is not any less reliable

prioritizing the lived experience is powerful

can follow up with feedback

true or false: research is either quantitative or qualitative

false

does not have to be either or

they compliment each other

gives both perspectives for full picture

true or false: all descriptive articles are qualitative

false

although qualitative research is descriptive, not all descriptive articles are qualitative

mixed methods study

if a study collects both qualitative and quantitative data using study designs for both

can quantitative data be presented in qualitative research?

yes!

qualitative research does not just analyze words

naturalistic inquiry

suggests that a phenomenon is only understood in context and from multiple perspectives (how you individually experience things)

inductive reasoning

data are collected and conclusions are reached based on this information

not going in with hypothesis

constructivism

our understanding of the world is "constructed"

no truth to reality, based on everyone's experience

research questions in qualitative research

hypotheses are avoided (want to discover new info, more focus on individual experience)

words that imply discovery

smaller sample sizes --> not designed for generalizability

purposive sampling

participants and settings are selected for a purpose or a specific reason

finding patients that can share in-depth experience

snowball sampling

initial participants are asked to recruit additional participants from their own networks -- helpful to have

naturalistic observation

observing events and phenomenon as they would occur in their natural or real-world setting

environment is key!

interview

one-on-one

led by a trained interviewer

semi structured interview guide with open-ended questions

main way to collect info

focus groups

multiple persons interviewed at once (usually 6-8 people)

led by a trained facilitator

semi-structured focus group guide to facilitate discussion

get dynamic interaction b/w group participants

artifacts

analysis of objects that provide information about the subject of interest

may also review documents or archives

photovoice

participants document research themes via photos or video

field notes

most unobtrusive way of recording and observing

focus on watching and listening

describe what is seen and researcher's impressions

participant observation

researcher engages with the participants in their naturally occurring activities

embedding themselves and actively participating

coding

1st task in data analysis

analyzing narrative data to find concepts to help make sense of what is going on

look for patterns or relationships, what is surprising, how data relates, and consider any inconsistencies

EVOLVING PROCESS

maintain a codebook

codes

smallest units of text that represent a theme

can be phrases or words

content analysis

systematic technique to draw inferences and coding textual material

involves counting frequency of words in text to analyze meanings and relationships of certain words, themes or categories within data

is data analysis time consuming?

hell yeah

given the amount of of detailed data typically produced with qualitative research

analysis by hand and software (NVIVO, atlas. ti, QDA miner

grounded theory

individual responses contribute to understanding theoretical relationships that can explain behavior

theory derived is "grounded" in the data

iterative process

what is the key distinguishing feature of grounded theory

simultaneous data collection and theory development

constant comparison

additional data --> new relationships

phenomenology

seeks to explain how events and circumstances influence perspectives and behaviors

phenomenology premise

lived experiences and meanings are only knowable to those who experience them

phenomenology researcher goal

elicit rich descriptions of study participants' experiences and meaning attributed to them

understanding the lived experience

phenomenology data collection

observation and lengthy discussion style interviews to get inside the experience of the participant and describe it as close as possible to how it is actually perceived

participatory action research

approach used to confront pressing social problems

combines research efforts and active intervention within single project

involves MULTIPLE STAKEHOLDERS as EQUAL PARTNERS

participatory action research process

look, think, act

begins with problem identification

participatory action research advantages

results are more likely to be trusted and used by consumers and professionals

results in new knowledge and change in practices, organization, policies, etc.

participatory action research challenges

creating community trust and sense of ownership of project and findings

ethnography

roots in anthropology

purpose is to describe a group of people, their behaviors, and/or their culture

thick description

cultural interpretation via an "insider's point of view"

relies on several sources of data

primary source is long term engagement via participant observation

critical appraisal

systemic process to identify the strengths and weaknesses of a research article in order to assess the usefulness and validity of research findings

QUANT and QUAL article reviews

reporting guidelines

guidelines written for researchers to help improve the value of published health research

randomized trials guideline

CONSORT

systematic reviews guideline

PRISMA

clinical practice guideline

AGREE

observational studies guideline

STROBE

scoping reviews guideline

PRISMA ScR

qualitative research guideline

SRQR

hawthorne effect

presence of researcher in research site

what is the four criteria of trustworthiness?

credibility

transferability

dependability

confirmability

credibility

capturing the "true" nature of the phenomenon

marked by thick description, triangulation, member reflections/member checking, negative case analysis

similar to internal validity

transferability

can the findings be transferred to other situations

marked by purposive sampling and thick description

similar to external validity

dependability

relates to the consistency between the data and the findings

marked by audit trail and triangulation

similar to reliability

confirmability

strategies used to limit bias in the research

maintain the neutrality of the data not the researcher

by a reflective researcher or review by peers

marked by audit trail, triangulation, and reflexivity

similar to objectability

reflexivity

journal of reflections, keeping track of reactions to bring awareness to bias

how this may influence interpretation of data

triangulation

more than one source

negative case analysis

explains conflicts

member checking

confirm interpretation with others, including participants and colleagues

audit trail

documenting decisions

saturation

collection continues until saturation is reached -- fully emerged in information

code-recode

data analysis, recoded, then evaluate findings

peer debriefing

serve as alternative viewpoints

what are the two forms of triangulation?

multiple sources of data

multiple data analysts

nuremberg code

the voluntary informed consent of the human subject is absolutely essential

declaration of helsinki

this protocol should be submitted... to a specially appointed ethical review committee, which must be independent of the investigator, the sponsor of any other kind of undue influence

belmont report

identifies the ethical principles upon which the federal regulations for human subject protections are based

includes respect for persons, beneficence, and justice

common rule

label given to the federal policy for the protection of human subjects

US HHS 45 CFR Part 46

what are the three main goals of informed consent?

inform participant of required information

document the participant was informed

establish participant VOLUNTARY and AUTONOMOUS decision to participate

are all informed consents the same?

no

it is not the same between research and medicine

but many commonalities at first glance (exceptions for emergencies)

assent

a child's affirmative agreement to participate in research

7-17 years of age

must be written at appropriate reading level of the youngest subject in the age range and use simple terminology

short form

when there is a language barrier and an IRB's approved consent is orally translated in the subject's native language

what method can you use to make sure that informed consent is understood by the participant?

teach back method

qualitative interviewing

goal is to let participant talk

key informants - selected for potential data they can provide

what is the most popular kind of qualitative interviewing?

semi-structured

has interview guide but is flexible and allows for varied probes

interview guides

used for research data collection

open and able to flow with participant

interviewer may reword questions, drop/add questions, and change sequence

probing

asking more thorough questions to receive more context

transcription

provides a verbatim guide to audio recordings

reflects actual words, speech, and thought patterns of interviewee

should remain relatively unedited

what is the transcribing time to recording time?

5:1 ratio

what are the benefits/risks with summarizing a transcription?

benefit: time saving

risk: overlooking and then forgetting relevant information

what are the best practices for transcription?

accuracy

do not paraphrase

do not add additional information

include speech patterns and language (and, but, you know, like)

leave out fillers (um, ah)

how do you transcribe unfinished thoughts?

dashes to indicate falters or incomplete thoughts, rambling speech or unfinished sentences

DO NOT USE ELLIPSES

how do you transcribe an indecipherable statement?

uncertainty: [?]

cannot understand what was said: [unclear]

unsure what was said: [text?]

how do you transcribe something that is potentially offensive or embarrassing to interviewee and should be left out?

[conversation regarding xxx redacted]

can you take out off-topic or extraneous remarks by the interviewer?

yes

words like "yes," "sure," i see," can be removed if becomes disruptive

how do you transcribe non-verbal sounds?

include and note with square brackets [ ]

how do you transcribe numbers?

use of numerals as long as do not begin the sentence

if start sentence, write out

how do you transcribe names?

provide full name of an acronym if known in brackets [ ]

what is the best approach to transcription techniques?

combine technology with quality checking

trained transcribers

when does the majority of data analysis occur?

after data collection

specific approaches like the grounded theory might involve analysis during data collection

what is the general process of data analysis?

coding -> categories/relationships -> development of themes

intimacy with the data <3

detailed and repeated readings

not enough to skim and a transcript

"cherry-picking" parts of the data

picking quotes from data without fully understanding it

leader to weaker descriptions and analysis

categories

2nd task in data analysis

consider theories, common-sense expectations and stereotypes

ensuring codes are refined in categories

use patterns and relationships to provide underlying meaning among categories

themes

3rd task in data analysis

using categories and their relationships to develop themes that answer the research question and possibly apply to other situations

typically define with quotes

conceptual mapping

4th task in data analysis

visually displays resulting themes and sub-themes

not always used, depends on researcher and design

can include labeled concepts, linking words, and clear hierarchies

can convey relationships among different concepts within a map (word links, directional arrows, simple connector lines, or overlapping circles)