Unit 1-2

Research Methods in Psychology

1. Research Methods
2. Ethics in Research
3. Statistical Analysis
 True or False?

“一个人独处会胡思乱想，不开心，加重消极情绪，容易抑郁。”

你对自己的结论有信心吗？
 你有预测到这个结果吗？
Common sense 为什么不可靠?
Perceive patterns in random events

Overconfidence
Overestimating the accuracy of your beliefs and judgments.

Hindsight bias (I-knew-it-all-along phenomenon)

• the tendency to believe, after learning an outcome, that one would have foreseen it.
• e.g. After a couple breaks up, friends say “they were never a good match.”

These roadblocks lead us to overestimate the weight of

commonsense thinking. Scientific inquiry can help us sift
reality from illusion.
 PRACTICE QUESTIONS

“Monday morning quarterbacks” rarely act surprised about the outcome of weekend
football games. This tendency to believe they knew how the game would turn out is
best explained by which psychological principle?
a. Overconfidence
b. Hindsight bias
c. Common sense b
d. Illusory correlation
e. Random sampling
 How do we do research in psychology ?

• Perceiving the Question Does sleep boost memory?

When sleep deprived, people remember

• Gathering information and Forming Hypothesis less from the day before. Or?

Give study material to people before: (1)

• Testing a Hypothesis by Conducting Research a long night’s sleep, or (2) a shortened
night’s sleep, then test memory.

• Data analysis and Drawing Conclusions confirm or reject the hypothesis

Research Methods
• Descriptive methods
• Correlational methods
• Experimental methods
 1. Descriptive methods

Describe behaviors, often by using case studies, surveys, or naturalistic observations.

 Case studies
• Examines one individual or group in depth.
• A small group of people (as little as one)

Example:
Brain damage, Little Hans

Pros:
• Allow for examination of rare behavior (or a rare psychological disorder), often used
in clinical psychology.
• Provide a full, detailed picture of one case (in depth).
Cons: the findings cannot be generalized the larger group.
 Naturalistic observations
• Observing and recording behavior in naturally occurring situations.
• Without manipulating the situation.

Counting positive words in 504 million Twitter messages.

 Naturalistic observations

Pros: 让一些无法在实验环境下观察的情况有研究的可能性.
Cons:
• Observer bias——tendency of observers to see what they expect to see.
• does not explain behavior, just describes it.

Example: A funny finding naturalistic observations

We humans laugh 30 times more often in social situations than in solitary situations.
But why?
 Surveys
• A technique for obtain the self-reported attitudes or behaviors.
Example: questionaries or interviews

Pros: easy and fast

Cons:
• Wording effects (framing): How an issue is worded can significantly affect decisions
and judgments.
e.g. Gun safety laws or gun control laws?
• Social desirability bias: the tendency to give answers that reflect well upon them.
• Sampling bias: a flawed sampling process that produces an unrepresentative sample.
 Representative Sample

Population
• All those in a group being studied, from which samples may be drawn.

Random sample
• A sample that fairly represents a population because each member of the group has
an equal chance of being chosen.

e.g. If you wanted to survey Nankai secondary school students about students’
playing Genshin Impact……population? random sample?
• You might number the names in the general student listing and then use a random
number generator (or draw from a hat) to pick your survey participants.
2. Correlational methods
Associate different variables.
Correlation
• A measure of the extent to which two variables are related, and thus of how well
either variable predicts the other.

• Positive correlation
• Negative correlation

Positive correlation or Negative correlation?

1. The more children and youth used various media, the less happy they were with
their lives.
2. The longer children were breast-fed, the greater their later academic achievement.
3. The more income rose among a sample of poor families, the fewer psychiatric
symptoms their children experienced.
Cons: correlation does not equal causation (lead to).

e.g. 自尊与抑郁负相关，那么低自尊会导致抑郁症吗?
 Illusory correlation
Perceiving a relationship where none exists (occurring by chance).

• superstitious behavior: tend to perceive patterns in random events; the accidental

timing of rewards can produce superstitious behaviors (such as lucky numbers,
lucky jeans…).

• confirmation bias: search for the result what they already believed.
e.g. 有人认为不孕夫妇领养孩子后更可能怀孕；女权男权。

• regression toward the mean: the tendency for extreme or unusual scores or events
to fall back (regress) toward the average.
3. Experimental methods
Manipulate variables to discover their cause and effect.
Solitude as an Approach to Affective Self Regulation

Study1
 How do we measure “solitude”, “affective” ?

Operational definition
• explain how you will measure a variable.
• It is required for experiments: allows for replication for the research.
Replication（reliance ）
• Repeating the essence of a research study, usually with different participants in different
situations, to see whether the basic finding extends to other participants and circumstances.

• Good research is both valid and reliable.

Valid: the extent to which a test or experiment measures what it is supposed to.
Reliable: when it can be replicated; it is consistent.
 Practice

Try generate operational definitions for the following variables:

Happiness number of smiles a person makes in a period of time

Popularity ratings from peer group; number of school elections won

Good music number of weeks at the top of the charts; number of albums sold
 Independent, dependent and confounding variables
Experiments can show cause-effect relationship because of:
• manipulating one or more factors (independent variables) to observe the effect on
some behavior (dependent variable);
• controlling confounding variables.

independent variables: the treatment

dependent variable: the results
confounding variables (controlled variable): factors that may bias the results; need to be
controlled.
By controlling…… if…… then……
 Independent, dependent and confounding variables

In Study1, Independent, dependent and confounding variable ？

independent variables: solitude vs. social interaction

dependent variable: the change in high-arousal affect over time

confounding variables: 种族、性别、电子设备、沙发、噪音……

 experimental group and control group
experimental group
the group exposed to the experimental treatment (treatment of the independent variable).

control group
no treatment which sets the base line; serves as a comparison for evaluating the effect of
the treatment.

In Study1, experimental group and control group ?

experimental group : “the sit-alone group”
control group : “the social interaction group”
experimental conditioning: the condition in which they were instructed to sit alone
control conditioning: the condition in which they were engage in a social interaction
 Random assignment
• Assigning participants to experimental and control groups by chance, thus
minimizing preexisting differences between the different groups.
• Random assignment ensures causation (ensures experiment).

比如：如果参与实验的志愿者中有三
分之一的人会结巴，那么每组中约有
三分之一的人会结巴。年龄、态度和
其他特征也是如此。
How to control confounding variables ?
 1. Situation-relevant confounding variables

Controlled by equivalent situations

The situations into which the different groups are put must also be equivalent
except for the differences produced by the independent variable.
 2. participant-relevant confounding variables

• Preexisting differences between the different groups

Controlled by random assignment

• Social desirability
Controlled by single blind procedure
Experiment where only the participants are ignorant to who received the treatment.
天然的单盲
 3. placebo/expectation effect

• A reaction based on a person beliefs more than on

the stimulus provided. Experimental results caused
by expectations alone.

• Controlled by placebo method or blind procedure.

Placebo method:
control group: 真药
experimental group: 一颗维生素

social desirability vs. placebo/expectation effect

4. Experimenter bias (experimenter-relevant confounding variables)

• Tendency for researchers to treat members of the experimental and control groups
differently to increase the chance of confirming their hypothesis (unconsciously).

Controlled by a double-blind procedure

• Experiment where both the participants and the researchers are ignorant to who
received the treatment.
• 比如，请一个毫不知情的人当研究人员。
 double-blind procedure vs. single blind procedure

• Single-blind technique: more appropriate when the experimenter is using a variable

that makes blinding impossible (e.g., race, sex).

• Double-blind technique: more appropriate to eliminate experimenter bias OR both

experimenter and participant bias.

• Both techniques can eliminate participant bias, placebo/expectation effect …

random sample vs. random assignment

Process
• population
• random sampling
• Participants (subjects)
• random assignment
Experimental design
 Between subject design and within subject design

Between subject design: random assignment separates participants into control

and experimental groups.
Within subject design: each participant is exposed to both the control and
experimental condition.

“Study1”实验是哪一种？

如果要改成 within subject design，应该如何做？

想要对比10-30岁，年龄不同的人的心理和行为特点，如何实现？

Cross-Sectional Study and Longitudinal Study

 Cross-Sectional Study

Research that compares people of different ages at the same time.

e.g. 2022年9月15日15选取一组10-30岁的人，对比这群人之间的心理和行为特点。

Pros: easier, cheaper, avoid attrition problem of longitudinal studies.

Cons: be careful to avoid the effects of historical events and cultural trends.
 Longitudinal Study

Research that follows and retests the same people over time.

e.g. 2021—2042年追踪一组10岁儿童20年，研究这群人的心理和行为发展特点。

Pros: more appropriate for following the same people over time (the focus is over time).

Cons: attrition problem; high cost; long time

e.g.《人生七年》
 laboratory experiments and field experiments

laboratory experiments: 在实验室，环境可控更能做出因果关系;

field experiments: 在自然世界，与现实情况更相似。

区分：黑猩猩研究

养在实验室中教黑猩猩数学采集数据
——laboratory experiments

在大自然中观察记录黑猩猩对数字的敏感度
——Naturalistic observations

在大自然中教黑猩猩数学并记录
——field experiments
PRACTICE QUESTIONS

1. Which method should a psychology researcher use if she is interested in

testing whether a specific reward in a classroom situation causes students to
behave better?
a. Case study
b. Experiment
c. Survey
b
d. Naturalistic observation
e. Correlation
 PRACTICE QUESTIONS

2. Which of the following describes the placebo effect?

c
a. Students in art class are not told that their work will be evaluated for a scholarship.
b. Participants in an experiment do not know if they are in the experimental or control group.
c. Participants in a drug study are given an inert pill instead of the drug and behave as though
they were given the drug.
d. Only females are chosen for a study, even though the population included males.
e. Experimenters manipulate one variable in a study, but not any others.
 b
PRACTICE QUESTIONS

3. A journalism student is writing an article about her school’s new cellphone policy,
and she’d like to interview a random sample of students. Which of the following is
the best example of a random sample?
a. The writer arrives at school early and interviews the first five students who come
through the main entrance.
b. The writer pulls the names of five students from a hat that contains all students’ names.
She interviews the five selected students.
c. The writer asks her teacher if she can distribute a brief survey to the students in her
AP® Psychology class.
d. The writer passes out brief surveys to 50 students in the hall and uses the 18 surveys
returned to her as the basis of her article.
e. The writer asks the principal for the names of 10 students who have had their cell
phones confiscated for a day for violating the policy. She interviews these 10 students.
 PRACTICE QUESTIONS

a
4. Why is random assignment of participants to groups an important aspect of a
properly designed experiment?
a. If the participants are randomly assigned, the researcher can assume that the people in
each group are similar.
b. By randomly assigning participants, the researcher knows that whatever is learned from
the experiment will also be true for the population from which the participants were
selected.
c. Random assignment keeps expectations from influencing the results of the experiment.
d. If participants are not randomly assigned, it is impossible to replicate the experiment.
Operational definition
e. Statistical analysis cannot be performed on experiment if random assignment is not used.
 PRACTICE QUESTIONS

5. There is a negative correlation between TV watching and grades. What can we conclude
from this research finding?
a. We can conclude that this is an illusory correlation. e
b. We can conclude that TV watching leads to lower grades.
c. We can conclude that TV watching leads to higher grades.
d. We can conclude that the grades students get impact their TV watching habits.
e. We can conclude that a student who watches a lot of TV is likely to have lower grades.
Ethics in Research
 Why do we need animal research ?

Pavlov’s Classical Skinner’s Operant

benefiting animals Conditioning Experiments Conditioning Experiments

benefiting humans
 Animal Ethics (较少考)

APA developed strict guidelines about what animals and how animals can be used in
psychological research.

1. They must have a clear scientific purpose, answer a important scientific question.

2. Animals chosen must be best-suited to answer the question at hand.

3. They must care for and house animals in a humane way.

4. They must acquire animal subjects legally. Animals must be purchased from accredited
companies. If wild animals must be used, they need to be trapped in a humane manner.

5. They must design experimental procedures that employ the least amount of suffering
feasible.
Phillip Zimbardo’s Prison Experiment Human Ethics
Deindividuation: guards and prisoners

Stanley Milgram’ Obedience Experiments

Participants thought they were causing significant
harm or death to other participants.

Risk
 Human Ethics (标红常考)

1. Informed consent 知情同意书

Participants must know that they are involved in research and give their consent. (若是
未成年人，需告知家长及监护人)

2. Risk
Protect them from physical or emotional harm and discomfort.

3. Confidentiality/Anonymity 匿名/保密性
Data must remain confidential.
 Human Ethics

4. Protection from Deception

Deception must be justified. In some cases, it is necessary to deceive the participants
because the study wouldn’t work any other way. The participants have to be told after the
study exactly why the deception was important (debriefing).

5. Debriefing 事后告知
Offer results or counseling (explain the research afterward).

6. No Coercion/Voluntary Participation 自愿
Participants may withdraw from the study at any time.
 PRACTICE QUESTIONS

e
1. Which of the following is not an ethical principle regarding experimental research
on humans?
a. Researchers must protect participants from needless harm and discomfort.
b. Participants must take part in the study on a voluntary basis.
c. Personal information about individual participants must be kept confidential.
d. Research studies must be fully explained to participants when the study is completed.
e. Participants should always be informed of the hypothesis of the study before they agree
to participate.
Statistical Analysis
We can test our hypotheses and refine our theories in 2 ways.

• Descriptive Statistics: simply describe a set of data.

• Inferential statistics: infer from sample data the probability of something

being true of a population.
1. Descriptive Statistics
Descriptive Statistics

simply describe a set of data.

 Histogram
a bar graph depicting a frequency distribution.

如图，一家美国卡车制造
商提供了图表(a)，表明其
卡车的耐用性更高。

• 垂直比例与差异的关系

• 考过画图
（1）Measures of Center
集中趋势 Describe the center of a data set.
 Measures of Central Tendency

mean the arithmetic average of a distribution, obtained by adding the scores

and then dividing by the number of scores.
mode the most frequently occurring score(s) in a distribution.
median the middle score in a distribution; half the scores are above it and
half are below it.

862410 mode mean median ？

 Percentile

• the standing of one score relative to all other scores in a set of data.

• For example, if your SAT score is in the 85th percentile, then you scored
higher than 85 percent of the other test takers.
运动员A的六次比赛分数为： 1 9 2 9 0 3 Mean=4 Mode=9 Median=2.5

运动员B的六次比赛分数为： 4 6 5 3 2 4 Mean=4 Mode=4 Median=2.5

你觉得应该派谁参加奥运会？
 （2）Measures of Variation
离中/变异趋势 It helps to know something about the amount of variation in the data—
how similar or diverse the scores are, includes range and standard deviation.
 range

The gap between the lowest and highest scores.

A的六次比赛分数为： 1 9 2 9 0 3 Range=9

B的六次比赛分数为： 4 6 5 3 2 4 Range=4
 standard deviation

Standard deviation: a computed measure of

how much scores vary around the mean score.

Standard deviation 越小，变异更小，数据离

散性越小，更集中。

Variance= 𝑆𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝑑𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛2

Z scores: 𝑡ℎ𝑒 𝑑𝑖𝑠𝑡𝑎𝑛𝑐𝑒 𝑜𝑓 𝑎 𝑠𝑐𝑜𝑟𝑒 𝑓𝑟𝑜𝑚 𝑡ℎ𝑒
𝑚𝑒𝑎𝑛 in units of s𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝑑𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛.
某一分数与平均值相差几个标准差
standard deviation (了解一下，一般不会考计算)

哪组变异更小，数据分布更集中？

Z scores
Class B, score=70, Z score=(70-80)/5=-2
A class: a mean of 120 and a variance of 25.

B class: a mean of 120 and a variance of 49.

B A
（3）Skewed distribution and normal distribution

数据的分布形态
 心理统计解释“不要内卷”

大部分正常人的智商都处在平均值附近。

Normal distribution
 Normal distribution
• A symmetrical, bell-shaped curve that describes the distribution of many types of data;
• Most scores fall near the mean and fewer near the extremes.
• mode = median = mean，可以用（normal curve）表示

Standard deviation=15

68% scores: fall within one standard deviation

of the mean
95% scores: fall within two standard deviation
of the mean
99.7% scores: fall within three standard
deviation of the mean
 skewed distribution 偏态分布
a representation of scores that lack symmetry around their average value.

口诀：
左歪—正偏态，众数在最左；
右歪—负偏态，众数在最右；

Positive skewed distribution（正偏态） Negative skewed distribution（负偏态）

Mode<Median<Mean； Mean<Median<Mode;
超过50%数据在mean的左边，mean靠右。 超过50%数据在mean的右边，mean靠左。
下图是什么偏态？

Positive skewed distribution

（4）Correlation coefficient
 correlation coefficient
A statistical index of the relationship between two variables

r from -1.0 to +1.0 表示相关的强弱和方向

r=0 代表什么？
Which correlation is negative ?
Which correlation is stronger ?

A correlation coefficient = 0.7

B correlation coefficient = -0.2

C correlation coefficient = -0.9

D correlation coefficient = 1.3

 Scatterplot
A graphed cluster of dots, each of which represents the values of two variables.

斜率表示方向、分散程度表明相关强度
 Positive correlation or Negative correlation?

Correlation coefficient ? r =1 r=0 1< r <0

2. inferential statistics 推论统计

Study1 对参与实验的人是独处更有助于自我情绪调节，
会不会是偶然发生的事件？
 Inferential statistics

Data that allow one to generalize—to infer from sample data the probability of something
being true of a population.

When Is an Observed Difference Reliable?

✓ Representative samples are better than biased samples. 样本有代表性

✓ Less-variable observations are more reliable than those that are more variable. 数据稳定

✓ Sample size: more cases are better than fewer. 更多样本

 Statistical significance

A statistical statement of how likely it is that an result occurred by chance (whether the
findings can be applied to the population).

p value<0.05 Its occurring by chance are less than 5 percent, thus the data is significant.

• by chance, null hypothesis < 0.05

• Alternative hypothesis > 0.95

Null hypothesis: a treatment had no effect in an experiment.

Alternative hypothesis: the treatment did have an effect.
 PRACTICE QUESTIONS

1. A student is interested in knowing how widely the academic aptitude of college-

bound students varies at her school. Which of the following statistical methods
should she use to determine how much students’ SAT scores vary from the
school’s average SAT score?
a. Correlation coefficient
b. Mean
c. Median
d. Standard deviation
e. Range
PRACTICE QUESTIONS

2. In a perfectly normal distribution of scores, which of the following statements

is true?
a. The mean, median, and mode are all the same number.
b. The mode is equal to the standard deviation.
c. The scores are positively correlated.
d. The mean minus the mode equals the median.
e. There is a positive skew to the distribution of data.
作业：

1. 练习册1-2 周一7点前交到办公室

2. 复习本单元内容，下周单元测验
AP考纲
本节重点
AP考纲
本节重点
Thank you