1/80
RRL TO QUANTITATIVE
Name | Mastery | Learn | Test | Matching | Spaced |
---|
No study sessions yet.
literature review
To determine if a topic is worth studying, narrow the scope, identify gaps, and form a foundation for research design.
research topic
The subject matter of the study (few words/short phrase) that becomes the central idea to explore.
3 strategies for developing a topic.
(1) Draft a concise working title (≤12 words),
(2) Pose topic as a research question,
(3) Check feasibility (participants/resources).
Availability of participants; access to resources (time, data, software).
Adds to knowledge, elevates underrepresented voices, advances social justice, has broader appeal, aligns with career goals.
Adds to knowledge, elevates underrepresented voices, advances social justice, has broader appeal, aligns with career goals.
What questions decide if a topic should be researched?
Integrate (summarize),
Criticize (point flaws),
Bridge (connect topics),
Identify issues (highlight debates/gaps)
Cooper’s four formats for using literature
RRL in qualitative
may place lit at intro, separate chapter, or end (inductive)
RRL in qualitative
usually structured, separate chapter early to test theory.
Steps for conducting a literature review (7 steps).
1) Identify keywords,
2) Search databases,
3) Locate ~50 reports,
4) Skim/select central works,
5) Design literature map,
6) Draft summaries,
7) Assemble review and highlight themes/gaps.
literature map
A visual diagram that groups literature and shows where the proposed study fits—how it adds/extends research
abstract (5 components)
Problem/issue, purpose/aim, data collection (who/where/type), results/findings, practical implications/audience.
APA (most common in psych), Chicago; consistency in style (citations, headings, tables) is essential.
What are key style manuals to consider?
Definition of terms
Clarity, consistency, precision, scientific accuracy. Define when first used; quantitative usually has a separate section.
Broad syntheses → reputable journal articles → books → recent conference papers → dissertations → cautious use of online sources.
literature selection — order of importance
Use of theory- Quantitative
tests hypotheses from theory (often early).
Use of theory- Qualitative
may generate theory (grounded) or use a lens (ethnography).
Use of theory- Mixed
can both test and generate; may use social justice frameworks.
“theory” in quantitative research
An interrelated set of constructs/propositions specifying relationships among variables to explain phenomena.
Micro: small time/space (e.g., face work).
Meso: link micro and macro (organizations).
Macro: explain large aggregates/societies.
What are micro-, meso-, and macro-level theories?
Early section titled “Theoretical Perspective” or integrated where it best guides hypotheses and data collection.
common placements of theory in a paper
theoretical-perspective section include?
Name and origin,
application topics,
central propositions/hypotheses,
and how the theory applies to IV→DV in the study.
“theoretical lens” in qualitative research
An orienting perspective (e.g., feminist, critical race) shaping questions and analysis with often transformative aims.
Early integration of framework, community involvement, advocacy stance, addressing power/inequities, ethical inclusion across process.
Mixed-methods participatory-social-justice framework — key features?
nine core arguments in proposal
1) What readers need to understand,
2) What readers need to know,
3) What you propose to study,
4) Setting & people,
5) Methods,
6) Analysis plan,
7) Validation,
8) Ethical issues,
9) Preliminary practicability evidence
Franklin’s three-stage writing model
1) Outline,
2) Draft and rearrange,
3) Edit/polish sentences
writing habit
Write daily, small regular sessions,
track progress,
break tasks into manageable pieces,
share drafts with peers
Readability
Umbrella: core idea;
Big: specific ideas under umbrella;
Little: reinforce big thoughts;
Attention: keep reader engaged and organized.
ethical steps before study
Consider codes of ethics,
apply to IRB,
obtain permissions,
select unbiased site,
negotiate authorship for publication.
Ethical considerations during data collection
Identify beneficial problem,
disclose purpose,
avoid pressure,
respect cultural norms,
minimize disruption,
provide benefits,
avoid deception (or debrief),
respect power imbalances.
introduction
Create interest, state the problem, place study in context, reach audience, clarify research purpose/significance.
Deficiencies Model's five steps for introductions
1) State problem,
2) Review past research,
3) Identify deficiencies (gaps),
4) Show significance/audience,
5) State purpose.
research problem
A narrative hook,
clear problem statement,
justification (many refs),
alignment with method,
and clear scope.
Qualitative introduction
explore phenomenon, personal/first-person style, use when concept immature. (Explore)
Quantitative introduction
explain relationships, objective style, theory-focused. (Explain)
Mixed study introduction
combine both—depends on emphasis. (Expand)
Start broad (related topics), then narrow to your focus—use the inverted triangle approach.
What to do when literature is scarce?
purpose statement
A sentence/paragraph conveying why the study will be conducted and what it intends to accomplish (aim/objective).
addressing the problem in an intro
Give broad summaries (categories/trends), position your study in dialogue, prioritize empirical studies, and use inverted triangle if literature is sparse.
qualitative purpose statement
Central phenomenon, participants, site, strategy of inquiry, action verbs (explore/understand), nondirectional language, delimit scope.
quantitative purpose statement
State theory/model, identify IVs/DVs (and order them IV→DV), specify strategy (survey/experiment), participants and site
mixed methods purpose statement
Overall intent, info about both strands, type of mixed design (convergent/explanatory/exploratory sequential), and rationale for mixing
Convergent design
collect qual + quant concurrently and compare.
Explanatory sequential design
quant first, then qual to explain (and vice-versa)
Qualitative research questions: central vs subquestions
1–2 central broad questions; 5–7 subquestions for details/interview prompts; open-ended wording; indicate participants and site.
quantitative research questions/hypotheses
Focus on variables and relationships; order IV before DV; base on theory; write clear null (H₀) and alternative (H₁) hypotheses; specify direction if directional.
quantitative questions: descriptive, relationship, comparison
Descriptive: frequency/distribution.
Relationship: association controlling for mediators.
Comparison: group differences in DV.
mixed methods research questions
Separate sections for quantitative questions/hypotheses, qualitative questions, and the mixed-methods integration question (how results inform each other).
Surveys
Describe trends, attitudes, opinions; test associations; answer descriptive, relationship, and predictive questions.
survey forms of data collection
Mail, telephone, Internet, personal interviews, group administration. Internet is common but has trade-offs.
Cross-sectional: one time point.
Longitudinal: multiple time points to observe change/predict.
Cross-sectional vs longitudinal surveys
Population: full group of interest;
sample: subset studied. Describe identification and access (frames)
population vs sample
sampling designs
Single-stage (direct sampling),
multistage/cluster (sample groups then individuals)
sampling methods
Random (equal probability),
systematic (every Xth after random start),
nonprobability/convenience (availability)
stratification
Divide population by known traits and sample proportionally to ensure subgroup representation—improves representativeness.
power analysis
Estimates sample size to detect effects; requires effect size, alpha (Type I, often .05), beta/power (Type II, often .20/power .8). Done before data collection
instruments
Name of instrument used, source, whether original/modified, sample items, scales, reliability (Cronbach’s α), and validity evidence (content, predictive, construct)
Cronbach’s α around .7–.9 is optimal.
What α range is considered acceptable for internal consistency?
pilot test
Before study. To check content validity, clarity, format, internal consistency, and participant fatigue; specify number of testers and revision plans.
survey instrument
Cover letter, demographics, attitudinal/behavioral/factual items, closing instructions; appropriate scaling (Likert vs categorical)
1) Advance-notice letter;
2) Survey sent 1 week later;
3) Follow-up postcard 4–8 days after;
4) Final personalized mail to nonrespondents 3 weeks later.
Total ~4 weeks.
four-phase mailed survey administration timeline
data analysis plan
1) Report responses (n, %),
2) Assess response bias,
3) Descriptives (means, SDs), handle missing data,
4) Scale scoring & reliability,
5) Inferential statistics (justify tests),
6) Present/interpret results (tables/figures).
response bias
Wave analysis, respondent vs nonrespondent comparisons
inferential reporting elements
Test statistics, p-values, effect sizes, and confidence intervals; justify test choice based on measurement scale and distribution.
results
Relate findings to research questions/hypotheses and prior studies, discuss theoretical/practical implications, limitations, alternative explanations, and future research.
experiment
Systematic manipulation of IV(s), control of extraneous variables, and (ideally) random assignment to infer causality.
experimental method plan
Participants & design, procedure, measures/instruments, manipulation checks, random assignment, control vs treatment groups.
pre-experimental designs
One-shot case study (A X O), One-group pretest–posttest (A O₁ X O₂), Static group comparison (A X O / B O)
quasi-experimental designs
Nonequivalent control-group, interrupted time-series (single-group or control-group versions)
Examples of true experimental designs
Pretest–posttest control-group (R O X O / R O O),
Posttest-only control-group (R X O / R O),
Solomon four-group
Single-subject design example
A–B–A design: baseline (A), treatment (B), withdrawal (A) to observe change/return to baseline.
participant recruitment description
How recruited (flyers, email), inclusion/exclusion criteria, compensation, selection process
study quasi-experimental
When there is no random assignment or only partial randomization.
matching and random assignment used for
To achieve group equivalence and reduce selection bias (random assignment best for internal validity).
validity threat
Internal, external, statistical conclusion, and construct validity threats.
internal validity threats (participant-related)
History, maturation, regression to mean, selection, mortality (attrition)
treatment-related internal threats.
Diffusion of treatment, compensatory rivalry, resentful demoralization.
procedure-related internal threats.
Testing effects (pretest influences posttest), instrumentation changes.
External validity threat
Person (sample not representative), Setting (artificial environment), Time (results specific to time/context)
statistical conclusion validity
Threats from low power, violations of assumptions, leading to incorrect inferences about statistical significance
construct validity
Whether variables are properly defined and measured—poor operationalization weakens conclusions.