Research Methods Review

hindsight bias

The tendency to believe, after learning an outcome, that we would have foreseen it.

overconfidence

The tendency to think we know more than we do.

What does it mean to perceive order in random events?

In our natural eagerness to make sense of an unpredictable world, we are prone to perceive patterns.

What is a theory?

an explanation using an integrated set of principles that organizes observations and predicts behaviors or events

What is a hypothesis?

a testable prediction, often implied by a theory

What is an operational definition?

A carefully worded statement of the exact procedures (operations)

used in a research study. 

For example, human intelligence may be

operationally defined as what an intelligence test measures.

How can we test hypotheses and refine theories?

descriptive methods, correlational methods, experimental methods

What are three descriptive methods?

case studies, naturalistic observation, surveys and interviews

What is a case study?

A descriptive technique in which one individual or group is studied in depth in the hope of revealing universal principles.

Strengths of the case study method

→ Allow for examination of rare or unusual behavior.

→ Provide a large amount of qualitative data.

→ Suggest directions Lfor further study.

Limitations of the case study method

→ Atypical case studies can be misleading.

→ Results from one study may not be generalizable to the larger group.

→ Cannot determine cause and effect.

What is naturalistic observation?

A descriptive technique of observing and recording behavior in

naturally occurring situations without trying to manipulate or

control the situation.

Strengths of the naturalistic observation method

→ Subjects behave “normally” outside of a lab setting.

→ Data collection is unobtrusive (doesn’t disturb the subject).

Limitations of the naturalistic observation method

→ Independent variable cannot be isolated.

→ Cannot determine cause and effect.

→ Observations by researchers may be subjective.

What is a survey?

a descriptive technique for obtaining the self-reported attitudes or behaviors of a particular group, usually by questioning a representative, random sample of the group.

Strengths of the survey method

→ able to take a “quick pulse” of people’s beliefs, behaviors or opinions

→ able to include many cases

Limitations of the survey method

→ response bias

→ wording effects can skew the outcomes 

→ acquiring a random sample is difficult

→ cannot determine cause and effect

What is a representative sample?

a subset of a population that accurately reflects the characteristics of the demographic it represents

population

anyone or anything that could possibly be selected to be in the sample

random sample

A sample that fairly represents a population because each member

has an equal chance of inclusion.

correlation

a measure of the extent to which two factors vary together, and

how well either factor predicts the other.

positive correlation

Two sets of data tend to rise or fall together

negative correlation

One set of data rises while the other falls

What is a scatterplot?

A graphed cluster of dots, each of which represents the values of two variables. The slope of the points suggests the positive or

negative direction of the relationship between the two variables.

The amount of scatter suggests the strength of the correlation.

correlation coefficients

→ can range from -1.0 to +1.0

→ reveals the extent to which two things relate.

→ The closer the score gets to +1 or -1, the stronger the correlation.

→ The (+) or (-) sign before the number indicates whether

the correlation is positive or negative.

illusory correlations

Perceiving a relationship where none exists Or perceiving a stronger-than-actual relationship.

What is regression to the mean?

The tendency for extreme scores or events to fall back toward the average.

random assignment

Assigning participants to experimental and control groups

by chance, thus minimizing preexisting differences between the different groups.

single blind procedure

The participants in the study are uninformed about the treatment, if any, they are receiving; Controls for subject response bias and placebo effect.

double blind procedure

The participants and the researcher are uninformed about which group receives the treatment and which does not; Controls for experimenter and subject bias as well as placebo effect.

What is the placebo effect?

A placebo is an inert treatment…like a pill without any medication

inside; The placebo effect causes experimental results simply from

expectations or assumptions that medication is being taken.

What is a confounding variable?

A factor other than the factor being studied that might influence a study’s results. E.g. age, sex, intelligence, ethnicity

What is experimental validity?

The extent to which a test or experiment measures or predicts what it is supposed to.

What ethical guidelines safeguard animal research?

1. The research must have a clear scientific purpose answering a specific important scientific question

2. The animals must be cared for and house in a humane way

3. The animal subjects must be acquired legally/purchased through accredited companies or if wild must be trapped in a humane manner

4. The experimental procedures employed must be designed to cause the least amount of suffering feasible

What ethical guidelines safeguard human research subjects?

1. Informed consent- Participants must know that they are involved in research and give their consent.

2. Coercion- Participation in the experiment must be voluntary

3. Anonymity/Confidentiality- Participants’ privacy must be protected

4. Risk- Participants cannot be placed at significant mental or physical risk

5. Debriefing- Participants must be told the purpose of the study and provided with ways to contact the researchers about the results of the study

What is descriptive statistics?

Numerical data used to measure and describe characteristics of groups. Includes measures of: central tendency (the mean, median and mode) and measures of variation (range and standard deviation).

mean

The mathematical average of a set of numbers. Add the scores and divide by the number (N) of scores.

median

The middle score in a distribution. Arrange scores from highest to lowest with half of the data above and half below this number.

mode

The most frequently occurring data point in a distribution.

What is a skewed distribution?

In a skewed distribution, most of the scores or data fall on one side of the scale and there are very few scores on the other side.

How does an outlier skew a distribution?

When one data point is extremely different from the others, this is called an outlier and can skew the results.

what is the range?

the difference between the highest and lowest scores in a distribution.

standard deviation

a computed measure of how much scores vary around the mean score.

To calculate the standard deviation:

Step 1: finCalculated the mean of your data

Step 2: calculate the difference between each data point and the mean

Step 3: Square each of these differences, add them up, and divide by the number of data points (minus one for a sample) to find the variance.

Step 4: Take the square root of the variance to get the standard deviation.

What is the normal curve (or a normal distribution)?

a symmetrical, bell-shaped curve that describes the distribution of many types of data; most scores fall near the mean and fewer and fewer near the extremes.

Characteristics of the normal curve

→ 68% of scores fall 1 standard deviation from the mean

→ 95% of scores fall 2 standard deviations from the mean

→ 99% of scores fall 3 standard deviations from the mean

descriptive statistics

→ describe a population or data set

→ use measures of central tendency

→ use measures of variation

inferential statistics

→ examine relationships between variables in sample

→ allow us to infer/predict trends based on data taken from a sample of a population

What is a “z score”?

a statistical measure that shows how many standard deviations a raw score is away from the mean of a dataset

Formula to calculate z score:

Z = (score - mean)/SD