Variables and Measurement Scales Notes - Study Guide

**Variables** 1. **Definition**: - A variable represents a category, object, event, or characteristic (e.g., age, gender, intelligence). - Variables must have **two or more levels** to allow meaningful differentiation. 2. **Types**: - **Quantitative Variables...

**Variables** 1. **Definition**: - A variable represents a category, object, event, or characteristic (e.g., age, gender, intelligence). - Variables must have **two or more levels** to allow meaningful differentiation. 2. **Types**: - **Quantitative Variables**: Have numeric properties (e.g., test scores). - **Categorical Variables**: Use numbers as placeholders (e.g., gender: male = 1, female = 2). **Measurement Scales** 1. **Nominal**: - Categories without numeric meaning (e.g., gender). 2. **Ordinal**: - Rank order with no equal intervals (e.g., Olympic medals). 3. **Interval**: - Equal intervals but no true zero (e.g., temperature in Celsius). 4. **Ratio**: - Equal intervals with a true zero (e.g., weight, time). - Simplified categorization for the class: - **Categorical Variables**: Includes nominal and ordinal. - **Continuous Variables**: Includes interval and ratio. **Special Cases** - Scales like Likert (e.g., \"Strongly Agree\" to \"Strongly Disagree\") are sometimes treated as **continuous** depending on the analysis. **Descriptive Statistics** 1. **Purpose**: - Summarize sample characteristics (e.g., age, gender breakdown). - Provide measures of central tendency and data spread. 2. **Key Measures**: - **Central Tendency**: - Mean: Average (for continuous data). - Median: Middle score (for ordinal, interval, ratio). - Mode: Most frequent value (for all types). - **Spread**: - Range: Highest score - lowest score. - Standard Deviation: Average distance from the mean. 3. **Choosing the Measure**: - Mean is suitable for symmetric data. - Median is better for skewed data (e.g., income distribution). **Using SPSS for Descriptive Statistics** 1. **Steps for Categorical Variables**: - Use frequencies and percentages (e.g., gender distribution). 2. **Steps for Continuous Variables**: - Calculate mean, standard deviation, min, and max. Week 3\ **Summary of the Lecture** This week focuses on understanding **variables** in developmental research, how to measure them, and evaluating the effectiveness of those measurements. Key topics include: 1. **Types of Variables**: - **Independent Variables (IV)**: These are manipulated or predetermined variables hypothesized to cause an outcome (e.g., the social context in an experiment involving friends versus strangers). - **Experimental IVs**: Actively manipulated by researchers (e.g., assigning exercise levels to participants). - **Subject Variables**: Naturally occurring attributes, such as age, gender, or diagnosis. - **Situational Variables**: Environmental or contextual factors, like classroom density or community resources. - **Dependent Variables (DV)**: These are the outcomes or effects measured in response to the IV (e.g., memory skills or self-esteem). 2. **Operational Definitions**: - Variables must be explicitly defined in terms of measurable criteria or procedures to ensure clarity and consistency in research. - Example: \"Aggression\" could be defined as the number of fights in school, scores on a personality scale, or parental reports of behavior. 3. **Challenges in Measurement**: - Some variables, like aggression, gender, or compliance, require subjective or multi-faceted measurement methods. - Researchers often use multiple methods (e.g., parent reports and behavioral observations) to create a comprehensive understanding and ensure results align across different approaches. 4. **Practical Applications**: - Operational definitions are crucial for clear communication of findings and ensuring comparability across studies. - Ethical and cost considerations often shape how variables are measured. 5. **Examples**: - **Experiment with Sharing**: Children's sharing behavior is observed based on their condition (friend vs. non-friend). - **Exercise and Memory**: Randomly assigned exercise levels are studied for their impact on older adults\' memory. - **Social Media and Self-esteem**: Measurement-based study exploring how social media use influences adolescent self-esteem. **Key Takeaways:** - Clear definitions of variables are critical to valid and replicable research. - Operational definitions help bridge abstract constructs and measurable outcomes. - Combining multiple measures often leads to a more reliable and thorough understanding of complex constructs. The lecture discusses different types of measurement methods in developmental and psychological research, their applications, advantages, and disadvantages. Here\'s a summary: **1. Self-Report Measures** - **Definition**: Participants (or proxies like parents) report on their behaviors or perceptions. - **Types**: - Oral (interviews) - Written or electronic (surveys) - Closed-ended (e.g., Likert scales) and open-ended questions. - **Advantages**: - Inexpensive, direct, and feasible. - Best for assessing thoughts, feelings, or personality traits. - **Disadvantages**: - Potential for dishonesty, bias (e.g., social desirability), or misinterpretation. - Requires participants to understand questions, which can be challenging for children or diverse cultural groups. **2. Behavioral Measures** - **Definition**: Direct observation of behaviors. - **Examples**: Observing children's helping behaviors or aggression. - **Preparation**: - Operationally define variables. - Develop a coding system and plan for ethical interventions. - Decide on live or recorded observations. - Train research assistants. - **Advantages**: - Objective and can capture non-verbal behaviors. - Feasible in naturalistic settings. - **Disadvantages**: - Reactivity (participants change behavior if observed). - Potential bias in coding. - Time-intensive (training, piloting, data collection, coding). **3. Physiological Measures** - **Definition**: Measures physiological responses like heart rate, cortisol, or brain activity (e.g., EEG, MRI). - **Advantages**: - Objective and difficult to fake. - **Disadvantages**: - Expensive and requires specialized equipment and expertise. - Potentially intrusive, causing participant hesitation. - Data interpretation can be complex. **4. Multimethod Studies** - **Definition**: Combines multiple measurement techniques (e.g., self-reports, observations, physiological measures) to study a single construct. - **Advantages**: - Provides a comprehensive view of the variable. - Increases confidence in the reliability of findings. - Detects differences across measurement methods (e.g., self-report vs. physiological data). **Conclusion** Each measurement type has strengths and weaknesses. Combining methods in a multimethod study is considered the gold standard for ensuring reliability and capturing diverse aspects of the construct being studied. **Summary: Measuring Variables in Research** **Purpose of Measurement** - Assign numbers to behaviors/events to reflect the characteristic or construct being investigated. - Ensures reliability, validity, and minimizes reactivity. **1. Reliability** **Definition:** Consistency or stability of a measure, ensuring it reflects the true score with minimal measurement error.\ **Importance:** Reliable measures ensure data accurately reflects constructs, avoiding spurious results. **Ways to Improve Reliability:** - **Use Multiple Items:** More items reduce error (e.g., measuring intimate partner violence with multiple items vs. one). - **Careful Procedures:** Train assistants and phrase questions effectively (positively worded items work better). - **Types of Reliability:** - **Internal Consistency:** Items in a measure should correlate (measured via Cronbach's Alpha). - **Test-Retest Reliability:** Stability over time, using repeated measures. - **Split-Half Reliability:** Correlation between two halves of a measure (e.g., odd vs. even items). - **Interrater Reliability:** Agreement among observers, improved by clear operational definitions. **2. Validity** **Definition:** The extent to which a measure accurately captures the intended construct. **Aspects of Validity:** - **Convergent Validity:** Measure correlates with other similar measures. - **Discriminant Validity:** Measure does not correlate with unrelated constructs. **Threats to Validity:** - **Reactivity:** Awareness of being measured alters behavior. **3. Reactivity** **Manifestations:** - **Social Desirability:** Participants alter responses to appear favorable. - **Observer Presence:** Participants change behavior when they know they're being observed. - **Physiological Reactivity:** Instruments (e.g., electrodes) can trigger nervous or altered responses. **Minimizing Reactivity:** - **Neutral Questions:** Distract participants from study focus. - **Warm-Up Period:** Allow participants to adapt to the study environment. - **Unobtrusive Measures:** Conceal observers to reduce behavioral changes. **Takeaways** Reliable and valid measurements are fundamental for accurate research results. By addressing reliability, validity, and reactivity, researchers can ensure that their data represents true participant behaviors and constructs. WEEK 4 Here's a summary of the lecture on writing a research article: **Key Objectives** 1. Communicate clearly through writing and presentations. 2. Critically evaluate research and carry out projects from start to finish. **Structure of a Research Article** - **Hourglass Shape**: Start broad, get specific, then broaden again. - **Introduction**: Broad context leading to specific research questions and hypotheses. - **Methods & Results**: The most specific and detailed sections. - **Discussion**: Broaden again to discuss implications related to broader concepts. **Writing Principles** - Focus on **accuracy** and **clarity** over style. - Use a **logical organization**---create an outline to ensure flow. - Write simply with direct, straightforward language. **Tips for Clear Writing** 1. **Main Points and Evidence**: Each paragraph should focus on one main idea supported by evidence. 2. **Logical Progression**: Use transition words (e.g., "therefore," "similarly") to connect ideas. 3. **Consistent Terminology**: Avoid substituting synonyms; consistency enhances clarity. 4. **Stay Relevant**: Cite studies directly related to your hypothesis and findings. 5. **Formal Language**: Avoid contractions, slang, or filler words. **Editing for Precision** - Omit unnecessary words and sentences while retaining essential details. - Avoid jargon unless necessary; define terms on first use. - Ensure consistent tense (e.g., past tense in the methods section). **Common Errors** - **Grammar**: Proofread carefully, paying attention to plural forms (e.g., \"data\" is plural). - **Word Choice**: Avoid errors like "while" instead of "although." **APA Format for Student Papers** 1. **Title Page**: Includes title, author, institution, course, instructor, and due date. 2. **Abstract**: Brief summary including aims, hypotheses, methods, findings, and a conclusion. 3. **Body**: Organized text starting on a new page with subheadings (if needed). 4. **References**: List sources for readers to locate cited works. 5. **Tables/Figures**: On separate pages with headings. 6. **Appendices**: Include SPSS outputs or supplemental materials. **Final Thoughts** - Scientific writing is linear and straightforward, requiring attention, revision, and time. - Aim to make your writing clear, concise, and easy for the reader to follow. Here's a summary of the lecture on writing a research article: **Key Objectives** 1. Communicate clearly through writing and presentations. 2. Critically evaluate research and carry out projects from start to finish. **Structure of a Research Article** - **Hourglass Shape**: Start broad, get specific, then broaden again. - **Introduction**: Broad context leading to specific research questions and hypotheses. - **Methods & Results**: The most specific and detailed sections. - **Discussion**: Broaden again to discuss implications related to broader concepts. **Writing Principles** - Focus on **accuracy** and **clarity** over style. - Use a **logical organization**---create an outline to ensure flow. - Write simply with direct, straightforward language. **Tips for Clear Writing** 1. **Main Points and Evidence**: Each paragraph should focus on one main idea supported by evidence. 2. **Logical Progression**: Use transition words (e.g., "therefore," "similarly") to connect ideas. 3. **Consistent Terminology**: Avoid substituting synonyms; consistency enhances clarity. 4. **Stay Relevant**: Cite studies directly related to your hypothesis and findings. 5. **Formal Language**: Avoid contractions, slang, or filler words. **Editing for Precision** - Omit unnecessary words and sentences while retaining essential details. - Avoid jargon unless necessary; define terms on first use. - Ensure consistent tense (e.g., past tense in the methods section). **Common Errors** - **Grammar**: Proofread carefully, paying attention to plural forms (e.g., \"data\" is plural). - **Word Choice**: Avoid errors like "while" instead of "although." **APA Format for Student Papers** 1. **Title Page**: Includes title, author, institution, course, instructor, and due date. 2. **Abstract**: Brief summary including aims, hypotheses, methods, findings, and a conclusion. 3. **Body**: Organized text starting on a new page with subheadings (if needed). 4. **References**: List sources for readers to locate cited works. 5. **Tables/Figures**: On separate pages with headings. 6. **Appendices**: Include SPSS outputs or supplemental materials. **Final Thoughts** - Scientific writing is linear and straightforward, requiring attention, revision, and time. - Aim to make your writing clear, concise, and easy for the reader to follow. This lecture focuses on crafting the **Method Section** of a research paper, providing essential guidelines and tips. Key points include: 1. **Purpose and Structure**: - The Method section includes **Participants**, **Measures**, and **Procedure** subsections. - Its goal is to detail the study's design and execution for reproducibility and validation. 2. **Participants Section**: - Describe the sampled population, recruitment methods, response rate, and eligibility criteria. - Include relevant demographic information and note IRB approval. 3. **Measures Section**: - Clearly outline the tools and methods (e.g., questionnaires, observations) used to collect data. - Define variables operationally to ensure clarity. - Provide details on measures\' scales, reliability (e.g., Cronbach's Alpha), and any modifications made. 4. **Procedure Section**: - Describe each step of the study, particularly differences by experimental groups, using past tense. - Include interrater reliability for coding (e.g., kappa values or correlations). - For content analyses, highlight coding processes and reliability steps. 5. **Writing Tips**: - Use concise, scientific language and complete sentences. - Avoid personal pronouns; instead, write in an objective tone (e.g., "Aggression was assessed using..."). - Reference APA-style papers for guidance on structuring and phrasing. 6. **Additional Advice**: - Proofread for clarity, consistency, and grammatical accuracy. - Avoid jargon unless necessary and define terms upon first use. The lecture emphasized that writing a Method section requires attention to detail and consistency to maintain scientific rigor and clarity. Week 5 This transcript provides an in-depth explanation of sampling methods and measurement techniques in developmental research, offering key insights into probability and non-probability sampling, their uses, benefits, and limitations. Here\'s a summary of the major points discussed: **Types of Sampling** **Non-Probability Sampling** - **Definition:** The probability of any particular member of the population being chosen is unknown. - **Types:** - **Convenience Sampling:** Collects participants easily accessible to the researcher. - **Quota Sampling:** Reflects the composition of subgroups within the population without random selection. - **Challenges:** Limited generalizability and potential bias. - **Why Use It?** - Affordability. - Feasibility for researchers. - Effective for studying relationships between variables rather than prevalence rates. **Probability Sampling** - **Definition:** Every member of the population has an equal chance of being chosen. - **Types:** - **Stratified Sampling:** Divides the population into subgroups (strata) and samples from each. - **Cluster Sampling:** Selects entire pre-existing groups or clusters (e.g., geographic areas). - **Multi-Stage Cluster Sampling:** Combines listing and sampling at multiple stages. - **Benefits:** - Higher generalizability. - **Challenges:** - Costly and logistically complex. **Evaluating Sample Quality** - **Sample Size:** Larger samples are better for generalization. Small samples risk sampling error. - **Response Rate:** The percentage of participants completing the survey impacts bias. - **Sampling Bias:** Occurs when sampling procedures overlook key population groups. **Case Study: Harlow's Attachment Research** - **Objective:** Investigated whether infant attachment depends on food or comfort. - **Findings:** Monkeys preferred the comfort of a cloth mother over a wire mother, even when the latter provided food. - **Generalizability:** - Limited external validity due to artificial conditions. - Contributed valuable theoretical insights into attachment. The discussion also highlighted that the choice of sampling method depends on research goals. For instance, while non-probability sampling is less generalizable, it is often sufficient for exploring relationships or patterns. On the other hand, probability sampling is essential when generalizing findings to a broader population. Let me know if you\'d like help unpacking any specific section or applying these concepts! This lecture covers **experimental methods** in research, focusing on the distinction between correlational and experimental designs. Here\'s a summary: **Key Concepts:** 1. **Research Goals**: - Scientists study the relationships between variables to describe, explain, and predict behavior. 2. **Two Approaches**: - **Correlational Design**: Observes natural behavior without manipulating variables to identify relationships. - **Experimental Design**: Manipulates the **independent variable (IV)** to observe its effect on the **dependent variable (DV)**, allowing causal conclusions. 3. **Experimental Method**: - Directly manipulates the IV to create different conditions (e.g., treatment vs. control groups). - Uses **random assignment** to ensure participant equivalence and control extraneous variables. - Assesses all participants on the same DV to isolate the IV's effect. 4. **Example**: - **Research Question**: Does viewing violent films increase aggression in children? - **Correlational Design**: Surveys parents on children's film exposure and aggression levels. - **Experimental Design**: - IV: Watching a violent (e.g., *Power Rangers*) vs. nonviolent (e.g., *Winnie the Pooh*) film. - DV: Number of punches/kicks in a play session. - Children are randomly assigned to groups, ensuring causation is attributed to the film type, not preexisting traits. 5. **Advantages of Experimental Design**: - Allows for causal statements. - Ensures control over extraneous variables. 6. **Limitations**: - **Artificiality**: Controlled settings may not reflect real-world behavior. - **Ethical Concerns**: E.g., withholding treatments in health studies may be unethical. - **Unrandomizable Variables**: Some factors (e.g., temperament, parenting style) cannot be randomized but remain important for research. In summary, experimental methods, particularly through **randomization and control**, are powerful for establishing causality but come with ethical and practical limitations. The **results section** of a research article is a concise description of your analyses, distinct from the discussion section, where findings are interpreted. Here\'s a summary: **Key Components of the Results Section:** 1. **Introduction to the Results**: - Restate research questions and hypotheses to remind readers of your predictions. - Specify the statistical test used (e.g., t-test, F-test, correlation) and justify its selection. 2. **Presenting Findings**: - Summarize findings in **words** and include **statistical results** (e.g., p-values, means) using **APA format**. - Present results in the same order as hypotheses from the introduction. - Write the narrative portion (word-based explanation) first, ensuring clarity, then add numerical results. 3. **Descriptive Statistics**: - Include mean, standard deviation, or frequencies to provide context for your data. - Descriptive statistics can be in-text or in a table. Use a table if: - There's a large volume of data. - Writing results becomes repetitive. 4. **Tables and Figures**: - For complex findings, you may present data visually in tables or figures (more common in advanced papers like Paper 2). 5. **What to Avoid**: - Do not interpret or discuss findings in detail---this belongs in the discussion section. - Refrain from linking findings to external literature or theoretical implications. **Decision on Placement:** - **Descriptive statistics** for study variables can be placed in the method section if they describe measures. - Statistics offering additional context should be included in the results section. The focus of the results section is on clearly and concisely reporting what was found, without delving into the broader implications. **Summary of Week 6 Content** This week\'s focus is on understanding **statistical tests** used in developmental research, emphasizing correlational designs. The learning goals include: 1. Grasping principles, designs, and methods of developmental research. 2. Applying appropriate statistical tests for various research designs. 3. Communicating findings effectively through written and oral presentations. 4. Demonstrating computer and information literacy to execute research projects. **Key Topics Covered:** 1. **Correlational vs. Experimental Designs**: - **Correlational Designs**: Researchers observe naturally occurring behaviors without manipulating variables. Examples include self-reports, observations, or analyzing public records. - **Experimental Designs**: Researchers manipulate independent variables to observe effects on dependent variables. 2. **Types of Variable Relationships**: - **Positive Linear Relationship**: Both variables increase or decrease together (e.g., anxiety and depression). - **Negative Linear Relationship**: One variable increases while the other decreases (e.g., alcohol consumption and GPA). - **No Relationship**: Variables vary independently (e.g., preference for a TV show and intelligence). - **Curvilinear Relationship**: A mix of positive and negative relationships, creating a U-shaped or inverted-U graph (e.g., anxiety levels and exam performance). 3. **Strength of Relationships**: - Measured with a correlation coefficient (ranging from -1 to 1). - Positive or negative signs indicate direction; magnitude indicates strength. - Pearson's correlation coefficient (r) is typically used, reported in APA format. 4. **Limitations of Correlational Designs**: - **Correlation Does Not Imply Causation**: - **Directionality Problem**: It's unclear which variable influences the other (e.g., hostile parenting vs. child misbehavior). - **Third Variable Problem**: An external factor (e.g., socioeconomic status) might explain the relationship. 5. **Examples of Third Variable Problems**: - A study linking parental involvement to academic performance might overlook socioeconomic status as a confounding variable. - An herbal antidepressant study may show mood improvement due to unrelated factors like weather or belief in the remedy. 6. **Clarifying Terminology**: - \"Correlation\" in this context refers to **correlational designs** rather than specific statistical tests. Correlational designs may use various tests (e.g., ANOVA, chi-square) but remain non-experimental, meaning causation cannot be inferred. This week emphasizes critical thinking about study designs and the limitations of drawing conclusions from correlational data. **Summary and Notes on Inferential Statistics** **Key Concepts in Inferential Statistics** 1. **Purpose**: Inferential statistics are used to: - Make inferences about a population based on a sample. - Test hypotheses. 2. **Choice of Statistical Test**: - Depends on the research question and variable types/levels of measurement. - Requires understanding continuous vs. categorical variables. **Quick Guide to Statistical Tests** - **Two Categorical Variables** → **Chi-Squared Test** - **One Categorical (2 Levels) + One Continuous** → **T-Test** - **One Categorical (2+ Levels) + One Continuous** → **ANOVA** - **Two Continuous Variables** → **Correlation** **P-Value** - **Definition**: The probability of findings being due to chance. - **Threshold**: - Typically set at **p \<.05** (less than 5% chance). - Smaller thresholds (e.g., **p \<.01**) increase confidence in findings but reduce sensitivity to smaller effects. - **Arbitrariness**: The.05 standard is convention, not an absolute rule. - **Interpretation**: - Smaller p-value = Greater confidence findings are not due to chance. - P-values don't indicate importance or size of effect, just confidence level. **Chi-Squared Test (χ²)** - **Use**: To test relationships between two categorical variables. - **Goal**: Determine if the distribution of one variable depends on the other. - **Steps in SPSS**: - Analyze → Descriptive Statistics → Cross Tabs. - Specify rows/columns for variables. - Check: - Statistics: Chi-Squared. - Cells: Observed and Expected Counts, Row/Column Percentages. **Example Outputs and Interpretation** 1. **Significant Results**: - **Example**: \"Children who preferred to play alone were more likely to have difficulty making friends (33.3%) compared to those who did not prefer to play alone (3.8%).\" - APA Style Reporting: Include χ² value, degrees of freedom, sample size, and p-value. 2. **Non-Significant Results**: - Simply report: \"No significant difference between groups (e.g., p =.27).\" **Visualization** - **Bar Graphs**: - Help clarify percentages and relationships between groups. - Example: Comparing difficulty in making friends by preference for playing alone. **Other Notes** - **Effect Sizes**: - Important to consider, especially in cases where smaller effects are meaningful. - P-value and effect size often work together to provide a fuller understanding. - **APA Style Guidelines**: - Use italics for statistical symbols. - Round values to two decimal places. **Practical Application** - For significant findings, include counts, percentages, and the statistical test result in your report. - For non-significant findings, omit detailed group-level comparisons. **Notes on Running Chi-Squared Tests in SPSS** **Steps to Perform Chi-Squared Analysis in SPSS** 1. **Access Cross Tabs**: - Go to **Analyze** → **Descriptive Statistics** → **Cross Tabs**. 2. **Variable Selection**: - Assign your variables: - **Row**: One variable of interest (e.g., Parenting Style). - **Column**: Another variable of interest (e.g., Child\'s Age). - Ensure variables are assigned to **both rows and columns** to avoid errors. - Right-click in the Cross Tabs menu to choose whether to display **Variable Names** or **Variable Labels**. 3. **Handling Variables**: - If using a typically continuous variable (e.g., Age), categorize it appropriately (e.g., \"Young Children\" vs. \"Teenagers\"). 4. **Setting Options**: - Under **Statistics**: - Check the box for **Chi-Squared Test**. - Click **Continue**. - Under **Cells**: - In **Counts**, check: - **Observed**. - **Expected**. - In **Percentages**, check: - **Row**. - **Column**. - Click **Continue**. 5. **Visualization**: - Enable **Cluster Bar Charts** to generate a visual representation of the results. 6. **Execute Analysis**: - Hit **OK** to run the analysis. - Alternatively, hit **Paste** to generate a **Syntax File**: - Save the file for later use. - Re-run the analysis by highlighting the specific syntax and clicking the **Green Play Button**. **Tips:** - **Save Syntax Files**: These allow you to re-run the analysis easily without re-entering options. - **Use Bar Charts**: Cluster bar charts provide an easy-to-interpret visual summary of relationships between variables. - **Categorical Variables**: Ensure all variables in a chi-squared analysis are categorical, even if continuous data needs to be grouped. **Notes on Performing and Interpreting T-Tests in SPSS** **Purpose of a T-Test** - **Objective**: Determine if there's a statistically significant difference in the means of a continuous variable between two groups of a categorical variable. - Example 1: Does GPA differ between males and females? - Example 2: Do children using picture books score higher on a literacy test than children watching videos? **Key Components** - **One Continuous Variable**: (e.g., GPA, test scores). - **One Categorical Variable** (with two levels): (e.g., Gender: Male vs. Female). **Steps to Perform an Independent Samples T-Test** 1. **Access the T-Test Functionality**: - Go to **Analyze** → **Compare Means** → **Independent Samples T-Test**. 2. **Assign Variables**: - Place the **continuous variable** (e.g., GPA) into the **Test Variable** box. - Place the **categorical variable** (e.g., Gender) into the **Grouping Variable** box. 3. **Define Groups**: - Tell SPSS how groups are coded: - Example: If \"1\" represents males and \"2\" represents females, define the groups as \"1\" and \"2\". - Adjust values if groups are coded differently (e.g., \"3\" and \"4\"). 4. **Set Options**: - Go to **Options** to ensure a **95% confidence interval** is selected. 5. **Run or Save the Analysis**: - Click **OK** to generate output. - Alternatively, click **Paste** to save a **Syntax File** for future use. - Highlight the T-Test syntax and click the **Green Play Button** to re-run the analysis. **Interpreting T-Test Output** 1. **Group Statistics Table**: - Provides the mean, standard deviation, and sample size (N) for each group. - Use this table for detailed reporting in results. 2. **Independent Samples Test Table**: - Focus on the **first row** for the following: - **T value** (italicized and small). - **Degrees of Freedom**. - **P-value**: - If **P \< 0.05**: Statistically significant difference exists. - If **P ≥ 0.05**: No significant difference. **Writing Results** - **Statistically Significant Example**: - *\"An independent samples T-Test showed significant differences in vocabulary test scores for children who did and did not have difficulty making friends, T(df) = value, P = value. Children with difficulty making friends had lower scores (M = value, SD = value) compared to children without difficulty (M = value, SD = value).\"* - **Non-Significant Example**: - *\"There was no significant difference in brattin scores based on gender, T(df) = value, P = value. Girls (M = value, SD = value) and boys (M = value, SD = value) did not differ significantly.\"* **Tips** - **Output Tables**: Always refer to the Group Statistics and Independent Samples Test tables for accurate reporting. - **Syntax Files**: Use syntax files for reusability and consistency in analyses. - **Focus on P-Value**: Determines whether to reject or accept the null hypothesis. **Summary and Notes: ANOVA (Analysis of Variance)** **Purpose of ANOVA** - **When to Use:** To test if there are differences in means across **more than two groups** or categories. - Example 1: Does GPA vary by class year (freshman, sophomore, junior, senior)? - Example 2: Does a literacy test score differ between children reading picture books, watching videos, or listening to podcasts? - **Difference from T-Test:** - T-test compares two groups. - ANOVA compares **three or more groups**. **Steps to Perform ANOVA in SPSS** 1. **Navigation:** - Go to **Analyze → Compare Means → One-Way ANOVA**. 2. **Input Variables:** - **Dependent Variable (Continuous):** Enter into the *Dependent List* box (e.g., GPA, test scores). - **Independent Variable (Categorical):** Enter into the *Factor* box (e.g., marital status, birth order). 3. **Options:** - Select **Descriptive Statistics** and **Means Plots**. - Under **Post Hoc**, choose the **Tukey test** for pairwise comparisons if the ANOVA shows significance. 4. **Run or Save Analysis:** - Click **OK** to generate results or **Paste** to save the syntax for later reruns. **Interpreting ANOVA Output** 1. **Descriptive Statistics Table:** - Provides mean, standard deviation, and group sizes (N) for each group. - Report these in your write-up for context. 2. **ANOVA Table:** - Key elements: - **Degrees of Freedom (DF):** Between groups and within groups. - **F-Statistic:** Shows the ratio of variance explained by the model to unexplained variance. - **P-Value:** Determines significance. - **Significance Decision:** - If **P \< 0.05**, there is a significant difference between at least two groups. - If **P ≥ 0.05**, no significant differences exist. 3. **Post Hoc Tukey Test (if significant):** - Identifies which specific groups differ from each other. - Look at **P-values** in the significance column: - Significant if **P \< 0.05**. - Example: \"Only children had significantly higher scores compared to middle and youngest children, but not compared to firstborns.\" 4. **Means Plot:** - Visual representation of group means to assist in interpreting differences. **Writing Results** 1. **If Significant:** - Start with a summary of the ANOVA. - \"A one-way ANOVA revealed significant differences in GPA across marital status groups (F(3, 452) = 3.87, p =.02).\" - Include findings from the post hoc test. - \"Only children (M = 5.3, SD = 0.8) scored significantly higher than middle (M = 4.2, SD = 1.0) and youngest children (M = 4.1, SD = 0.9).\" 2. **If Not Significant:** - Report only the ANOVA result. - \"A one-way ANOVA revealed no significant differences in GPA across marital status groups (F(3, 452) = 0.93, p =.738).\" - Skip post hoc test results. **Key Takeaways** - **ANOVA Tests Main Effect:** Determines if there is a significant difference among group means. - **Post Hoc Analysis:** Explains **where** differences exist (if any). - Use **descriptive statistics** and **mean plots** to supplement interpretations. - No significant ANOVA result means no need for post hoc analysis. **Example Write-Up** **Significant Result:** \"A one-way ANOVA revealed significant differences in children\'s emotional problems based on birth order (F(3, 452) = 3.87, p =.02). Post hoc Tukey tests indicated that only children (M = 5.3, SD = 0.8) scored significantly higher than middle (M = 4.2, SD = 1.0) and youngest children (M = 4.1, SD = 0.9), but not firstborns.\" **Non-Significant Result:** \"A one-way ANOVA revealed no significant differences in emotional problems based on birth order (F(3, 452) = 0.93, p =.738).\" Post hoc:\ **Example Write-Ups** **Example 1: Significant Differences** \"A Tukey post hoc test was conducted to determine which specific groups differed after a significant ANOVA result. The test revealed that only children (M = 5.3, SD = 0.8) scored significantly higher on the emotional problems scale than middle children (M = 4.2, SD = 1.0, p =.03) and youngest children (M = 4.1, SD = 0.9, p =.02). However, no significant differences were found between only children and firstborns (p =.07).\" **Example 2: No Significant Differences** \"Following the ANOVA, a Tukey post hoc test indicated no statistically significant differences between any of the groups (all p-values \>.05). This suggests that the observed group means do not differ significantly.\" **Key Components to Include** 1. **Group Pairs Compared**: Specify which groups were compared. 2. **Means and Standard Deviations**: Include descriptive statistics to provide context. 3. **P-Values**: Show significance level for each comparison. 4. **Plain Language**: Highlight findings in a reader-friendly way. **Correlation Test Summary and Notes** **Purpose of Correlation** - Determines whether two continuous variables are related. - The correlation coefficient (r) ranges from **-1 to 1**: - **Positive r**: Variables increase together (positive relationship). - **Negative r**: One variable increases as the other decreases (negative relationship). - **r = 0**: No relationship. **Strength of Correlation (Magnitude)** - **Strong**: \|r\| ≥ 0.7 - **Moderate**: 0.4 ≤ \|r\| ≤ 0.69 - **Weak to Moderate**: 0.1 ≤ \|r\| ≤ 0.39 - **No or Weak**: \|r\| ≤ 0.09 **Steps to Compute Correlation in SPSS** 1. Go to **Analyze \> Correlate \> Bivariate**. 2. Place the two variables of interest in the **Variable** box. 3. Ensure **Pearson\'s correlation coefficient** and **two-tailed test** are selected. 4. Click **OK** to generate output or **Paste** to save syntax for later. **Interpreting Correlation Output** - **Pearson's Correlation Coefficient (r)**: - Measures the strength and direction of the relationship. - **Significance Level (p-value)**: - If **p \< 0.05**, the correlation is statistically significant. - **Sample Size (N)**: - Indicates the number of observations used for the correlation. **Example Interpretation:** - Variables: Children's height and weight. - Results: r=.45r =.45, p\ 1. Takes place in participants\' natural settings (e.g., playgrounds). 2. Observes behaviors like sharing, aggression, peer group formation, and social interaction. 3. Purpose: Understand how individuals in a social/cultural setting live and interact naturally. 1. **Systematic Observation**: 1. Observations focus on *specific variables* and hypotheses. 2. Procedures are highly structured with coding systems in place. 3. Less emphasis on immersion compared to naturalistic observations. 4. Requires **clear operational definitions** for live coding. **Key Methodological Aspects of Observations** 1. **Coding Systems**: - Systems must be **simple and operationally defined** to ensure ease of use, especially during live observations. - Live coding differs from reviewing recorded behaviors, where observations can be paused and analyzed frame by frame. 2. **When to Observe**: - **Continuous Sampling**: Observing behaviors over extended periods for richer data. - **Time Sampling**: Observations at regular intervals (e.g., every few seconds or minutes). - **Event Sampling**: Observations focused on specific events or behaviors (e.g., during a holiday). - **Latency Recording**: Measures the response time between a stimulus and a behavior (e.g., how long it takes a student to sit after a bell rings). **Evaluating Observational Research** 1. **Interrater Reliability**: - Ensures two or more observers agree on observations. - Measured using indices like **ICC or Kappa**. - High reliability is expected (80%+ agreement). 2. **Reactivity**: - Reactivity occurs if the presence of the observer alters participants' behavior, creating a threat to internal validity. - **Ways to Reduce Reactivity**: - Use **unobtrusive methods** (e.g., concealed equipment like one-way mirrors or hidden recording devices). - Allow participants time to acclimate to the observer\'s presence. **Advantages & Disadvantages of Observations** **Advantages:** 1. **Objective Data**: No self-report biases. 2. **Standardized**: Uniform coding schemes across all observations. **Disadvantages:** 1. **Cost & Time**: Coding and equipment can be expensive and time-consuming. 2. **Reliability Issues**: Observer inconsistency can compromise results. 3. **Missed Data**: Observational schemes might not capture all important behaviors. **Case Example: Madson & Collins (2008) Study** **Study Purpose:** - Investigated the relationship between **adolescent interactions with parents at age 13** and **romantic partnerships at ages 20--21**. - Examined **expressive vs. collaborative processes**: - **Expressive Processes**: Emotional expressions (e.g., positive/negative affect, conflict). - **Collaborative Processes**: Conflict resolution and balancing personal and relationship needs. **Findings:** 1. **Expressive Processes**: - These were not predictive of romantic relationship behaviors later. 2. **Collaborative Processes**: - Observed parental collaboration predicted later **collaborative and expressive behaviors** in romantic partnerships. **Theoretical Implications:** - Findings align with attachment theory. - Parental collaborative processes (e.g., conflict resolution, problem-solving) are stronger predictors of later relationship functioning compared to emotional expression alone. **Final Insights** - Observational research offers insight into natural behaviors but requires careful methodological design. - Systematic observation allows researchers to examine specific variables using clear coding systems and structured procedures. - Reactivity and interrater reliability are two primary challenges that researchers mitigate with strategic observation methods and standardized coding. - Examples like **Madson & Collins'** study demonstrate real-world applications and the significance of observational techniques in predicting relationship development over time. **Summary of Lecture:** This final lecture focuses on **developmental designs**, **single subject designs**, and **professional development skills**. The main learning goals are: 1. Understand principles and methods of developmental research designs. 2. Improve written and oral communication of research findings. 3. Learn how to execute a complete research project from start to finish. **Key Concepts** **Developmental Designs** Developmental research examines **age as a primary variable** and explores how developmental changes occur over time. Methods of studying this include: 1. **Cross-Sectional Method**: - Involves studying different groups of people at different ages at one point in time. - Example: Comparing 6-year-olds, 9-year-olds, and 12-year-olds in 2009. **Advantages**: - **Time-efficient**: Data collection only happens at one point. - **Cost-effective**: Less time and financial resources needed. - Limits **historical events\' impact** because it\'s not spread across years. **Disadvantages**: - Focuses on **age differences**, not changes over time. - Cohorts (groups sharing common historical/cultural traits) may differ, influencing data. 2. **Longitudinal Method**: - Involves studying the same group of individuals over two or more time points. - Example: Measuring the same individual\'s college adjustment over two points in a semester. **Advantages**: - Observes **actual developmental change** over time. - Eliminates cohort effects (variability between age groups). **Disadvantages**: - **Time-consuming and expensive**. - **Attrition**: Participants may drop out of the study for various reasons. - Methods to reduce attrition: - Offering financial incentives. - Frequent communication to maintain participation. - Emphasizing the study\'s societal importance. **Other Challenges**: - **Historical Events**: Events (like elections or pandemics) could confound findings. - **Practice Effects**: Repeated assessments may alter responses due to familiarity with the process. - Measurement changes must remain developmentally appropriate across time points. **Example of Longitudinal Study:** The **Doney Longitudinal Study**: - A multi-decade study following 1,037 New Zealand children born between 1972--73. - Participants have been assessed every 2 years from birth through age 38. - Included physical, mental health, and relationship data. - The study has led to **1,150+ academic publications**. The example highlights the range of **longitudinal research designs**, which can span **years or even just hours** as long as repeated data is collected at multiple time points. **Single Subject Designs** The lecture briefly transitioned into **single subject designs**, likely focusing on their application for individualized, intensive research or interventions, particularly in educational or therapeutic contexts. **Final Notes** - **Developmental research** compares age differences (cross-sectional) or observes actual changes over time (longitudinal). - Cohorts, attrition, historical events, and practice effects are significant factors in interpreting findings. - Longitudinal methods allow deeper insights into change but require careful planning to minimize bias. Understanding these research designs is essential for exploring developmental changes, individual differences, and educational or psychological outcomes. **Summary & Key Notes** **Single Subject Designs** - **Definition**: A single subject design involves only one subject in a study. The subject could be an individual, a classroom, or even a community/city. - **Goal**: Establish a causal relationship between **independent variables (IV)** and **dependent variables (DV)**. - Example: Changes in behavior correspond to the presence of an IV (e.g., rewards). **Structure of Single Subject Designs** 1. **Baseline Period**: Behavior observed before the independent variable is introduced. 2. **Treatment Period**: Independent variable introduced (e.g., reinforcement such as praise or tangible rewards like stickers). 3. **Observation**: Behavior continues to be observed to assess changes. **Issue in Single Subject Designs** - **Problem**: Changes in behavior may not always be due to the IV (e.g., other random events could influence behavior). **Solution: Reversal Designs (ABA Designs)** - **Definition**: A *baseline-treatment-baseline* sequence. - **Structure**: - **A (Baseline)**: Observe behavior without treatment. - **B (Treatment)**: Introduce IV (e.g., praise). - **A (Baseline)**: Remove treatment and observe behavior again. - Example: Attempting to reduce hyperactivity by giving students extra playtime contingent on good classroom behavior. - **ABAB Designs**: Repeat the treatment and baseline periods multiple times to strengthen evidence of causality. **Why ABAB Designs Are Important** 1. **Increased Confidence**: Evidence is more reliable when the effect is observed multiple times. 2. **Reduces Coincidences**: Random behavioral shifts are less likely to occur across repeated ABAB cycles. **Ethical Concerns in ABA/ABAB Designs** - Withdrawing a treatment at the end of the experiment could harm participants, especially if the treatment proves beneficial. - ABAB design can end with the treatment phase (instead of withdrawing it) to improve ethics. **Limitations of Single Subject Designs** 1. **Small Sample Size**: Hard to generalize findings to larger populations (N = 1). 2. **Inferential Statistics Issue**: Statistical testing is not typically applicable with N = 1 studies; results are instead graphed or charted. 3. **Limited Variables**: Behavior must be observed systematically and respond predictably to manipulation. **Conclusion** Single subject designs are helpful for observing individualized effects of interventions but have limitations such as small sample size, ethical concerns, and lack of inferential statistical testing. They rely on systematic observations and visual analysis rather than traditional statistics. **Summary & Key Notes** **Professional Development Skills** As the course concludes, the focus is on **marketable professional development skills** that will be beneficial as students pursue careers in psychology, child and adolescent development, or related fields. **Most Marketable Skills** 1. **Communication Skills** - Key examples: Writing, speaking, presenting, active listening, giving/receiving feedback, and negotiating. - Essential for nearly all careers. 2. **Teamwork Skills** - Includes: Collaboration, honest communication, and responsibility. - Valued by employers across industries. 3. **Leadership Skills** - Involves: Active listening, dependability, giving/receiving feedback, and patience. - Important at all career levels, from contributing to projects to managing teams. 4. **Interpersonal Skills** - These involve building relationships, effective communication, and conflict resolution. - Employers value: Motivation, flexibility, and empathy. 5. **Problem-Solving Skills** - Key traits: Communication, decision-making, research, and resourcefulness. - Essential in managing challenges and contributing to organizational success. 6. **Open-Mindedness** - Willingness to learn, adapt to new methods, and maintain a positive professional demeanor. 7. **Work Ethic** - Employers prioritize employees who are dependable, self-motivated, and can meet deadlines independently. 8. **Learning Skills** - Adaptability and continuous learning are vital. - Examples: Collaboration, communication, and critical thinking. 9. **Self-Management Skills** - These include time management, organization, and self-motivation. - Help prioritize tasks, maintain productivity, and contribute to growth. 10. **Organizational Skills** - Critical for time management, productivity, and achieving goals. - Examples: Planning, attention to detail, conflict management, and critical thinking. 11. **Computer Skills** - Basic computer literacy is essential in most roles. - Advanced skills (if applicable) include proficiency in: - Word processing, spreadsheets, social media, data visualization, email communication. **Reflection** Students should critically evaluate how they\'ve developed these skills not only in this class but throughout their entire college career. Recognizing and articulating these skills will enhance resumes and prepare them for job opportunities.