PSY2013 Lecture 03 Notes

PSY2013 Research Methods in Psychology

Ch04 Sampling, Measurement, and Hypothesis Testing

---

Who to Measure – Sampling Procedures 测量对象 - 抽样程序

Samples vs. Populations 样本 vs. 总体

• Population 总体
  • 研究人员希望通过研究了解的一组个体。
• Sample 样本
  • 从总体中选择的个体群体，用于代表总体进行研究。

Sampling Procedure 抽样程序

Probability Sampling 概率抽样

• Simple Random Sampling 简单随机抽样

  • 总体中的每个成员都有相等的机会被选为样本成员，例如彩票。
  • 有时使用随机数生成器从总体中选择。

• Stratified Sampling 分层抽样

  • Proportions of important subgroups in the population are represented precisely in the sample.
  • 总体中重要子群体的比例在样本中被精确代表。
  • Example: 75% female; 25% male (2 strata).
  • 例如：75%女性；25%男性（2层）。

• Cluster Sampling 聚类抽样

  • Randomly select a cluster of individuals all having some feature in common.
  • 随机选择具有某些共同特征的个体群体。
  • Example: Sample 3/20 dormitory students who live on-campus, and random select from each dormitory.
  • 例如：抽样3/20住在校园宿舍的学生，并从每个宿舍中随机选择。

Non-Probability Sampling 非概率抽样

• Convenience Sampling 方便抽样

  • Select subjects who are available and convenient, e.g., “subject pool”.
  • 选择可用和方便的受试者，例如“受试者池”。
  • 受试者自愿参与。

• Purposive Sampling 目的抽样

  • Selecting specific individuals who meet certain criteria.
  • 选择符合特定标准的个体。
  • 例如：Milgram不使用大学生。

• Quota Sampling 配额抽样

  • Similar to stratified sampling, but non-random.
  • 类似于分层抽样，但非随机。

• Snowball Sampling 滚雪球抽样

  • 请求受试者让他们的朋友、家人和熟人参与。
  • 通常通过在线调查进行。

---

What to Measure – Varieties of Behavior 测量对象 - 行为的种类

Developing Measures from Constructs 从构念中发展测量工具

• Reaction Time 反应时间
  • One of psychology’s original methods.
  • 心理学的原始方法之一。
  • Example: Mental chronometry (Donders).
  • 例子：心理计时法（Donders）。

Expanded Notes 扩展笔记:

• Mental Rotation 心理旋转
  • Construct: Visual imagery.
  • 构念：视觉想象。
  • Measure: Reaction time.
  • 测量：反应时间。

Evaluating Measures 评估测量工具

• Reliability 信度

  • Repeatability and consistency of measures.
  • 测量的重复性和一致性。
  • Results from a minimum of measurement errors.
  • 测量误差最小化的结果。
  • Quantified using correlation coefficient.
  • 使用相关系数量化。

• Validity 效度

  • Measures what it is designed to measure.
  • 测量其设计测量的内容。

Expanded Notes 扩展笔记:

• Types of Validity 效度类型
  • Content Validity 内容效度: Measures broadly represent what the measure is supposed to measure.
    • 测量广泛代表其应测量的内容。
  • Criterion Validity 标准效度: The extent to which a measure is related to some criterion (outcome variable).
    • 测量与某些标准（结果变量）相关的程度。
    • Concurrent: How well the measure predicts present performance.
      • 现时效度：测量预测当前表现的效果如何。
    • Predictive: How well the measure predicts future performance.
      • 预测效度：测量预测未来表现的效果如何。
  • Construct Validity 构念效度: The extent to which the abstract construct can be inferred from the operational definition.
    • 从操作定义推断抽象构念的程度。

Scales of Measurement 测量尺度

• Nominal Scale 名义尺度
  • Category-defined, e.g., gender, ethnicity.
  • 类别定义，例如性别、种族。
• Ordinal Scale 顺序尺度
  • Rank order, e.g., order of finish in a test.
  • 排序，例如测试中的完成顺序。
• Interval Scale 间隔尺度
  • Order + equal intervals, e.g., temperature scales.
  • 顺序 + 相等间隔，例如温度刻度。
• Ratio Scale 比率尺度
  • Order + equal intervals + true zero point, e.g., weight.
  • 顺序 + 相等间隔 + 真实零点，例如重量。

Expanded Notes 扩展笔记:

• Example of Scales 尺度示例
  • Nominal: Gender (male, female).
  • Ordinal: Education level (high school, bachelor’s, master’s).
  • Interval: Temperature (Celsius).
  • Ratio: Height (cm).

---

Statistical Analysis 统计分析

Descriptive Statistics 描述统计

• Frequency Distribution 频数分布
  • Count or frequency for nominal/categorical data.
  • 名义/分类数据的计数或频率。
• Measures of Central Tendency 集中趋势测量
  • What scores are at the center of a distribution.
  • 分布中心的分数。
  • Mean, median, mode.
  • 平均数、中位数、众数。
• Measures of Variability 变异性测量
  • How spread out or dispersed scores are in a distribution.
  • 分布中分数的分散程度。
  • Range, standard deviation (SD), variance (VAR), interquartile range (IQR).
  • 范围、标准差（SD）、方差（VAR）、四分位距（IQR）。

Inferential Statistics 推论统计

• Inferential Statistics 推论统计
  • Makes inferences and predictions about a population based on a sample of data.
  • 基于样本数据对总体进行推断和预测。
  • Provide a way to test a claim or hypothesis using sample data.
  • 提供一种使用样本数据检验主张或假设的方法。

---

Hypothesis Testing 假设检验

Null Hypothesis 零假设

• Definition 定义
  • No relationship (“no difference”) between variables in the population is expected.
  • 预计总体中变量之间没有关系（“没有差异”）。

Alternative Hypothesis 备择假设

• Definition 定义
  • A relationship (“a difference”) between variables in the population is expected.
  • 预计总体中变量之间有关系（“有差异”）。

Expanded Notes 扩展笔记:

• Significance Testing 显著性检验
  • If the observed p-value is less than a predefined value (alpha, α), the null hypothesis can be rejected.
  • 如果观察到的p值小于预定义值（alpha，α），则可以拒绝零假设。
  • Common alpha level: α = 0.05.
  • 常见的alpha水平：α = 0.05。

Possible Errors 可能的错误

• Type I Error 类型I错误
  • Rejecting a true null hypothesis (false positive).
  • 拒绝正确的零假设（假阳性）。
• Type II Error 类型II错误
  • Failing to reject a false null hypothesis (false negative).
  • 未能拒绝错误的零假设（假阴性）。

Expanded Notes 扩展笔记:

• Power 功效
  • The chance of rejecting a false null hypothesis.
  • 拒绝错误的零假设的机会。
  • Power = 1 - β (where β is the probability of a Type II error).
  • 功效 = 1 - β（其中β是类型II错误的概率）。

---

Beyond Hypothesis Testing 超越假设检验

Confidence Intervals 置信区间

• Definition 定义
  • The range within which the population mean is likely to be found within a percentage of the time.
  • 总体均值可能在一定百分比时间内出现的范围。
  • Example: A 95% confidence interval tells us that we can be 95% confident that the interval captures the population mean.
  • 例子：95%的置信区间告诉我们，我们可以95%确定该区间包含总体均值。

Effect Size 效应量

• Definition 定义
  • Emphasizes the size of the difference between variables, not merely whether there is a difference.
  • 强调变量之间差异的大小，而不仅仅是是否存在差异。

Open Science, Pre-registration, and Replications 开放科学、预注册和复制研究

Expanded Notes 扩展笔记:

• Open Science 开放科学
  • Promotes transparency and accessibility in scientific research.
  • 促进科学研究的透明性和可访问性。
• Pre-registration 预注册
  • Registering study design and analysis plans before conducting the research.
  • 在进行研究之前注册研究设计和分析计划。
• Replications 复制研究
  • Repeating studies to confirm findings and ensure reliability.
  • 重复研究以确认发现并确保可靠性。

---

Summary 总结

• Various sampling procedures are available to find a representative sample from the population.
• 有多种抽样程序可用于从总体中找到具有代表性的样本。
• Statistics is the language used to communicate findings in the field of psychology.
• 统计是心理学领域中用于交流发现的语言。
• Hypothesis testing involves determining the likelihood that the observed results are due to chance.
• 假设检验涉及确定观察到的结果是由于机会的可能性。
• Beyond hypothesis testing, confidence intervals, effect sizes, and power are essential for understanding psychological phenomena.
• 超越假设检验，置信区间、效应量和功效对于理解心理现象至关重要。

---