pp2 stats

Communicating Findings

• Importance of communication in research: effectively share findings through various methods.

• Report types:

  • Written Report (e.g., lab report)

  • Research Reporting: quantitative, qualitative, mixed methods

  • Literature Reviews

  • Systematic Reviews

  • Theory Papers

  • Other types: commentaries, letters to the editor

• Presentation formats:

  • Oral Presentations

  • Posters

APA Publication Manual

• Importance of adhering to APA style (visit www.apastyle.org for guidelines).

• Features:

  • Better guidance for quantitative writing

  • Improved guidance for qualitative research in the 7th edition

• Write-up formatting: includes journal submission and reference formatting

• Most psychology journals require APA style; not mandatory for student work (especially in the UK).

Good Scientific Writing

• Key attributes:

  • Clear: published works may not exemplify clarity.

  • Precise: provide enough detail for replication.

  • Good grammar: use past tense, avoid florid language.

  • Data is plural (e.g., "data are not" vs. "data is not").

  • First-person writing varies; check expectations.

Parts of Quantitative Write-Up

• Common sections include:

  • Title

  • Abstract

  • Introduction

  • Method (with subsections)

  • Results

  • Discussion

  • Reference List

  • Appendix may be included if necessary.

Title

• Aim for 12 words or less: concise and informative.

• Example formats:

  • "The effect of X on Y."

  • "Hippocampal Processing of Ambiguity Enhances Fear Memory."

• Avoid unnecessary words like "an investigation into".

Abstract

• Write the abstract last: summarize the report briefly.

• Up to 120 words recommended by APA Manual; some journals may allow up to 200.

• Contains key information: problem, tested ideas, method, results, conclusion.

Introduction

• Effectively introduce the problem/topic of study.

• Summarize relevant research, gradually narrowing to specific hypotheses:

  • Begin broadly, zero in on specific hypotheses/RQ’s.

  • Conduct the V-shape introduction.

Method

• Detailed enough for replication and judging reliability of findings.

• Potential subsections:

  • Participants

  • Design

  • Apparatus (optional)

  • Materials (optional)

  • Stimuli (optional)

  • Ethics (optional)

  • Procedure

Results

• Present findings through clear prose:

  • Include statistics (e.g., to two decimal places).

  • Discuss inferential statistics distinctly according to the hypothesis/RQs.

• Include means/SDs if relevant; state significance without further interpretation.

• Visual aids (graphs/tables) can enhance comprehensibility; refrain from raw SPSS outputs.

Discussion

• Summarize and interpret findings for each hypothesis/RQ:

  • Have findings relationship to previous research.

  • Discuss implications for field/society and the strengths/limitations of the study.

  • Suggest future steps; conclude with a takeaway message.

Sharing Findings at a Conference

• A venue for researchers/practitioners to exchange ideas and research:

  • Commonly attended by postgraduates.

• Two primary aims:

  • Share research/practice (scientific programs).

  • Network effectively (social programs).

• Research sharing format:

  • Oral presentations

  • Posters

Basics of Experiments and Analysis

• An experiment is a controlled situation to test variable effects.

• Example research question: "What is the effect of caffeine on well-being?"

• Experimental manipulation identifies causal relationships (Does X cause Y?).

Experimental Designs

• Conditions set up to compare independent variable effects:

  • ID: Independent Variable, which may require a control condition for comparisons.

  • Assess outcome variables (dependent) across conditions.

Important Design Considerations

• Ensure the only differing factor between conditions is the IV.

• Control other variables to make a fair test; randomization is key.

Design Variations

• Varied conditions may assess multiple levels of the IV:

  • Minimum of two conditions required.

• Options include testing multiple IVs/DVs, and using either between-subjects or within-subjects designs.

• Counterbalancing mitigates order effects in within-subject designs.

Pre- and Post-Test Methodology

• Measures can assess conditions before interventions, offering insight into changes attributable to manipulation.

Experimental Manipulations

• Strategies to create variable levels:

  • Situational: Design scenarios akin to Asch studies.

  • Tasks: Vary between computer/paper methods, etc.

  • Instructions: Direct participant actions.

• Maintain consistency, manipulating only the targeted IV to avoid confounding results.

Experiment Locations

• Research can occur in labs (more control, less ecological validity) or field settings (greater ecological validity, but more uncontrolled variables).

• Consider creating a lab-like environment within field settings.

Example Experiment: Pavlov's Work

• Hypothesis: Utilizing new mnemonic methods can enhance memory performance.

• IV: Feet vs. no new mnemonic among participants.

• DV: Number of recalled syllables.

Experiment Variations

• Variation A: Between-subjects; subject assignment leads to a mnemonic or no mnemonic condition.

• Variation B: Within-subjects; all subjects undertake both conditions, incorporating counterbalancing for order.

Quasi-Experiments

• Conduct studies where random assignments aren't feasible (e.g., gender, age).

• Cannot establish causation but can explore relationships.

Randomized Controlled Trials (RCTs)

• Considered the gold standard for testing interventions (e.g., drug efficacy).

• Follow methodological rigor yet still face inherent issues;

  • Emphasize robust randomization and strong controls within a field context.

Analyzing Experimental Data

• Mostly involves ANOVA variations based on experimental design (within, between, mixed).

• Statistical significance (p < .05) implies confidence in IV effects.

Examples of Analyses

• Differences in experiments analyzed through standard methods including:

  • Chi-square tests for nominal data.

  • ANOVA for various designs.

Recap of T-Statistics

• Central tendency of samples helps establish comparisons against population means.

• Use t-statistic when population SD is unknown; it accounts for sample variability and degrees of freedom (df).

Significance and Reporting T-Tests

• Interpret results with appropriate significance levels; reject or do not reject null hypotheses accordingly.

• Properly format reporting results (e.g., t(18) = -3.00, p < 0.05).

Independent vs. Paired Samples t-test

• Differentiation based on sample conditions; independent for distinct groups and paired for repeated measures.

Practical Application in PSPP/Study Design

• Engage in practical work to conceptualize t-test applications:

  • Formulate simple research questions using t-tests to derive significant insights.

  • Adhere to specified conditions, even when normal distribution might not hold.