Psychology Quantitative Research Methods

Aim

the purpose of the study - indicates which behavior/mental process will be studied

Credibility

Refers to the degree to which the results of a study can be trusted to reflect the reality (closely linked to bias)

Generalizability

The degree to which the behaviors observed in one's research study would be representative of those found in the larger population

Constructs

Theoretical definition of a concept; not directly observable e.g. love, anxiety

Operationalising

Describes when a variable (construct) is defined by the researcher and a way of measuring that variable is developed for the research.

Sampling bias

When errors in gathering a sample result in an unrepresentative sample, compromising the validity of the research

Opportunity sampling

Convenience sampling; a sample of whoever happens to be present and agrees to participate

Opportunity Sampling Strength

Useful when financial resources are limited. In some studies, there may be reasons to believe that people are not that different

Opportunity Sampling Weakness

Generalization from opportunity samples is very limited because of the sampling bias

Representative sample

A sample that accurately represents a population, in terms of ethnicity, gender, etc. i.e all essential characteristics

Self-selected sampling

A sampling method made up of volunteers

Self-selected Sampling Strength

A quick and easy method to recruit participants while at the same time having wide coverage

Self-selected Sampling Weakness

Representativeness and generalization are limited

Random sampling

Type of sampling where every member of the target population has an equal chance of being selected -- "putting all the names in the hat" -- best way to obtain a representative sample.

Random Sampling Strength

If the sample size is sufficient, researchers may be certain that even unexpected characteristics are fairly represented in the sample.

Random Sampling Weakness

It is practically impossible to carry out truly random sampling, for example, the target population might be geographically dispersed.

Stratified sample

This type of sample draws random samples from each subgroup (ethnic, gender, etc.) within the target population -- representative, but expensive/time-consuming to gather

Stratified Sampling Strength

Allows researchers to control representativeness of some key characteristics without relying on chance. Useful when the researcher is certain about which characteristics are essential and when the sample sizes are not large.

Stratified Sampling Weakness

Requires more knowledge about characteristics of the target population; harder to implement.

Experiments

when the researcher manipulates the IV while maintaining strict control of environment; participants are randomly allocated/assigned to conditions, shows cause and effect relationship clearly

Correlation Studies

This has a focus on two variables (not termed IV and DV, instead referred to as covariables) as the hypothesis is not based on potential cause and effect

Bias

any varibales that threaten the internal validity of the study

Independent Variable

The variable that the researcher is looking to find the effect of, that he/she deliberately manipulates

Dependent Variable

The variable that is being measured after the manipulation of the independent variable

Confounding/Control Variables

Variable that is not expected, and therefore not controlled for, by the experimenter; could affect the validity of the study's findings

Independent Measure Design

Members of the sample are randomly allocated to one condition of the experiment

Repeated Measure Design

One sample of participants that receives each condition of an experiment

Matched Pairs

An independent samples design in which participants are not randomly allocated to conditions. Instead, they are ranked based on certain characteristics and paired together before being allocated to a certain group.

Independent Measure Design Strength

- No Order effects (learning, fatigue, boredom)
- No Demand characteristics
- Same test can be used and have multiple groups

Independent Measure Design Weakness

- Subject variables differ
-Worse stat tests because of variation between conditions
-More subjects are required

- Not much equivalency

Repeated Measure Design Strength

Subject variables are kept constant between conditions
-Better stat tests
-Fewer subjects are required

- Higher equivalency

Repeated Measure Design Weakness

Order effects (learning, fatigue, boredom) from same test
-Demand characteristics (can guess aim)
-Different tests needed

Matched Pairs Strength

Subject variables are kept more constant between conditions
-Better stat tests
-Order effects don't occur since subject only is in one condition
-Same test can be used

- Higher equivalency

Matched Pairs Weakness

Subject variables can't be perfectly matched
-Time consuming to find matches
-More subjects are required

Independent Measure Design Solution

• Random allocation of large groups

Repeated Measure Design Solution

• counterbalancing though it can be difficult

Matched Pairs Solution

• keeping the experiment as simple as possible

Variability

the extent to which participants are different

Equivalency

How similar groups are so that they can be compared without worrying about other confounding varibales

Validity

another way to discuss findings is to consider whether the research does what it claims to do

Reliability

the results can be replicated and it used in reference to experimental study and if another person does the same procedure, it should give the same results

external validity

Degree to which the findings of the study can be generalized to other contexts -- so, other people, other cultures, etc.

Ecological Validity

The degree to which the findings obtained in a lab experiment would be found in other settings, outside a controlled environment

Population validity

degree to which the study results can be generalized to and across the people in the target population due to the sample being representative of the population

construct validity

the extent to which variables measure what they are supposed to measure

internal validity

Degree to which study truly shows a cause-effect relationship between two factors (or whether some unaccounted-for "third factor" led to the results obtained)

Cross-cultural validity

Whether or not the findings of a research study would be found in/relevant to other cultures, or if it is ethnocentric

History Bias

Outside events that happen to participants in the course of the experiment.

Sampling Bias Counteraction

Random allocation into groups; sufciently large group sizes

History Bias Counteraction

Standardize experimental procedures as much as possible in all groups

Maturation Bias

The natural changes that participants go through in the course of the experiment

Maturation Bias Counteraction

Having a control group.

Testing Effects

The frst measurement of the DV may affect the second (and subsequent) measurements.

Testing Effects Solution

In independent measures designs there must be a control group, the same test and retest, but no experimental manipulation. In repeated measures designs, counterbalancing must be used

Instrumentation

Occurs when the instrument measuring the DV changes slightly between measurements, compromising standardization of the measurement process.

Instrumentation Solution

Standardize measurement conditions as much as possible

Regression to the mean

This becomes a threat when the initial score on the DV is extreme

Regression to the mean Solution

A control group with the same starting score

Experimental Mortality

Occurs when some participants drop out of the experiment.

Experimental Mortality Solution

Whenever possible, design experimental conditions in such a way that participants do not feel discomfort

Demand Characteristics

Occurs when participants understand the true aim of the experiment and alter their behaviour

Demand Characteristics Solution

Deception

Experimenter Bias

Occurs when the researcher unintentionally infuences participants’ behaviour and the results of the study

Experimenter Bias Solution

Using the double-blind design

Single Blind Design

Experimental procedure aimed at reducing demand characteristics; the participant does not know the aim or purpose of the experiment

Double Blind Design

Experimental procedure aimed at reducing researcher bias; neither the participants or the person conducting the experiment know the aim or the purpose of the experiment

Types of Experiments

Laboratory, Field, Natural, and Quasi

Laboratory Experiment Strength

cause and effect
-more control and accurate measurements
-greater ability to replicate

Laboratory Experiment Weakness

total control over all variables is not possible
-artificial conditions leads to lack of ecological validity
-biased

Quasi Experiment

Allocation into groups is done on the basis of pre-existing differences, for example, age, gender, cultural background, education, occupation.

Quasi Experiment Strength

- great ecological validity
-very little bias

Quasi Experiment Weakness

- hard to infer cause and effect b/c little control
-impossible to replicate exactly
-ethical problems

Natural Experiment

Experiment where researchers cannot randomly assign participants to control/experimental conditions, where IV is "assigned by nature"

Natural Experiment Strength

- great ecological validity
-very little bias

Natural Experiment Weakness

hard to infer cause and effect b/c little control
-impossible to replicate exactly
-ethical problems

Field Experiment

Experiment conducted in a more natural setting, outside a laboratory

Field Experiment Strength

-greater ecological validity since behavior occurs in own environment
-less bias

Field Experiment Weakness

- More bias and greater difficulty to control
-difficult to replicate and record data accurately
-ethical problems

Negative Correlation

when one variable increase and the other decreases

Positive Correlation

when both variables are affected in the same way

Curvilinear Bias

In calculating the correlation between two variables, we assume that the relationship between them is linear. Mathematically the formula of a correlation coefficient is a formula of a straight line. However, curvilinear relationships cannot be captured in a standard correlation coefcient.

Third Variable Bias

There is always a possibility that a third variable exists that correlates both with A and B and explains the correlation between them.

Spurious Bias

Spurious correlations are correlations obtained by chance.

Curvlinear Counteraction

If suspected, curvilinear relationships should be investigated graphically

Third Variable Counteraction

Consider potential “third variables” in advance and include them in the research study

Spurrious Counteraction

Results of multiple correlations should be interpreted with caution.

Size of Correlations

less than 0.10 → negligble

0.10 - 0.29 → small

0.30 - 0.49 → medium

0.50 - or larger → big

Statistical Significance

p = n.5 non significant

p= <0.5 significant

p = <0.1 very significant

p = < 0.001 highly significant

Informed Consent

participants have to be informed about the nature of the study and agree to participate

Deception

sometimes the researcher doesn't want the participants to know the exact aims so they mislead them slightly — it can cause less stress but has to be explained after the study

Protection from harm

no harm can be done to participants and it's not allowed to humiliate someone or force them to reveal private info

Debriefing

after the study, the true aims and purposes of the research must be revealed to the people and deception has to be justified and all participants should leave study without stress

Right to withdraw

participants should be told they have the right to leave the study at any time and they can ask for their data to be deleted/removed at the end of the study if they want to

Anonymity and Confidentiality

all info has to be handled carefully and kept private

Primary and Incidental Findings

Primary findings are those that are expected to be found and necessary for the study while incidental are those that come about without meaning to

TARGET POPULATION

the group of people to which the findings of the study are expected to be generalized

SAMPLE

the group of people taking part in an experiment / study

Null Hypothesis

states that IV will have no effect of the DV or that any change in the IV will be due to chance
-researcher wants to show cause and effect relationship
-we can never prove anything we can only disprove

Experimental Hypothesis

predicts the relationship between the IV and the DV—what we expect will come out of the manipulation of the IV

Hypothesis

A statement that leads to further investigation with an IV and DV