IB HL Psychology - Research Methods and Concepts

Quantitative data

usually numerical data (objective)

Qualitative data

rich data that is highly descriptive (subjective)

Aim of quantitative data

Parameter	Quantitative data
Aim	To characterize behaviour of large group of individuals

Focus of quantitative data

Parameter	Quantitative data
Focus	Behavioural manifestation (operationalized → making it measurable)

Data of quantitative data

Parameter	Quantitative data
Data	Number

Objectivity of quantitative data

Parameter	Quantitative data
Objectivity	More objective

Types of quantitative data

Parameter	Quantitative data
Types (research methods)	Experiment, quasi-experiment, correlational study

Aim of qualitative data

Parameter	Qualitative data
Aim	In-depth + rich understanding of case or phenomenon

Focus of qualitative data

Parameter	Qualitative data
Focus	human experiences, interpretations, meanings

Data of qualitative data

Parameter	Qualitative data
Data	Text

Objectivity of qualitative data

Parameter	Qualitative data
Objectivity	More subjective

Types of qualitative data

Parameter	Qualitative data
Type (research methods)	Observation, interview, focus group, case study

Thing you need to have when doing an experiment

• Construct validity

• Sample

• Standardization of the procedure

• Internal validity

Construct validity

to what extant are you measuring what you need to measure

Sample

Participants

Standardization of the procedure

the process of ensuring that all participants in a study or experiment experience the exact same conditions, materials, instructions + environmental factors

Internal validity

Are you controlling things that can affect the results

Objectivity

The results generalized are not influenced by attitudes, personal feelings, opinions + experiences → a way that potential sources of bias are minimized.

Controlled

In experimental methods, all variables are operationalized + control over extraneous variables are applied → can enable researchers to establish cause + effect

Extraneous/confounding variables

variables other than the IV that could potentially influence the DV + change study results

Falsifiable

For something to be scientific, it should be falsifiable, should be amenable to scientific testing, should generate hypotheses → each support/refute a theory → can lead to a reconsideration of a theory

Nomothetic approach

creating laws that allow things to be generalized

Advantages of the empirical method

• creates general laws

• preventative measures

• treatments + interventions

Example of an experiment with a control group

Practice effect

Getting better as you do more of it → e.g. doing it 2x

Demand characteristics

Participants able to guess aim + many of them changed their results

Types of demand characteristics

• Social desirability bias

• Expectancy effect

• Screw you effect

• Participants biases

Social desirability bias

Give a result that’s a socially acceptable response

Expectancy effect

Try to help them out by giving results needed for researcher

Screw you effect

Giving total opposite response to researcher.

Participants biases

Systematic way where participants' expectations, motivations, or behavior influence the study's outcome

Reactivity (qualitative)

Participants alter their behavior because they are aware of being observed

Placebo (non - active treatment) effect

Person experiences a real change in symptoms/behavior because they believe it's a treatment

Nocebo effect

Person experiencing negative effects caused by belief that something will harm them

Characteristics of a true/lab experiment

• There is an artificial setting → extraneous variables are controlled

• Participants are randomly allocated to minimize researcher + participant bias

• starts off with a hypothesis → gives it a sense of direction → in experiments, never prove, but try to disprove the null hypothesis (which is no significant difference → IV manipulated, no change in the DV)

• IV is manipulated

• DV is measured in a quantitative way

• Establishes causality

Characteristics of a field experiment

• Inability to control extraneous variables → takes place in the natural environment

• IV is manipulated

• Participants are not randomly allocated

• Establishes cause + effect

• Does not take place under controlled conditions, but in real-world environments

Characteristics of a natural experiment

• IV is not manipulated since the IV is naturally occurring

• Occurs in a natural environment

• Measures the impacts on the DV

• All natural experiments are a type of quasi-experiments

• Happens when we cannot ethically manipulate the IV

• Cannot establish causality → only correlation

Characteristics of a quasi experiment

• Allocation of participants isn't random

• Participants are grouped based on their characteristics of interest

• Researchers are measuring pre-existing differences in 1 variable

• Methods become correlational and cannot draw conclusions → not establishing causality

• IV cannot be manipulated

External validity

Generalize the results to other populations/settings → are you able to generalize beyond the specific conditions of the experiment → do the findings apply to the real world or just this specific study?

Pros + Cons of true/lab

Experiment	Strengths	Limitations
True/lab ↑ internal val. ↑ control of variables ↓ eco val.	Cause + effect between the IV + DV can be confidently inferred due to strict controls Well-documented procedures make true experiments easy to replicate → ensuring reliability → standardised High control over extraneous variables ensures the study isolates the IV's effect on the DV, enhancing int. validity.	Artificial nature of research contexts often means participants don’t show real-life behavior → reducing ecological validity Demand characteristics can arise, participants might change their behavior based on cues from the researcher or situation → affecting the study’s results True/lab experiments may be impractical/ unethical in certain situations → limits its use

Pros + Cons of field

Experiment	Strengths	Limitations
Field ↑ eco val. ↓ control of variables ↓ internal val. ↓ demand characteristics ↓reliability ↓ replicability	high ecological validity since it takes place in the participants natural environment more unlikely that they suffer from demand characteristics as they may be unaware of the aim	less control over extraneous variables so lower int. val → might weaken the establishment of cause + effect

Pros + Cons of quasi

Experiment	Strengths	Limitations
Quasi	allows researchers to conduct experiments on pre-existing groups → e.g. gender differences in anxiety → hence not manipulating the IV	Causality cannot be established → cannot be sure of the equivalence of comparison groups at the start of study → pre-existing groups may have unexpected confounding variables Allocation into groups is not random → leaves room for bias

Pros + Cons of natural

Experiment	Strengths	Limitations
Natural ↑ eco val. ↓ internal val.	Allows phenomena to be studied that couldn’t be replicated in a true experiment due to ethical + practical reasons Changes in the IV are naturally occurring + not manipulated by researchers → high in eco val.	many extraneous variables that researchers cannot control → low internal validity IV is not directly manipulated → cannot draw definite causality conclusions

Pros + Cons of correlational studies

Experiment	Strengths	Limitations
Correlational studies (for pilot studies → most time use correlational studies)	Measures the strength of relationship → can lead to further detailed research → saves money on time measuring variables that have an important impact on each other Little manipulation of variables needed → researchers just measure existing variables → ↑ eco val	cannot establish cause + effect → researchers can only suggest relationship between 2 Variables need to be operationalized (measurable) → may lead to lacking int. vali

Non - directional hypothesis (2-tailed)

There will be a significant difference between the number of digits recalled that will be read aloud and that those will be read sub-vocally (A ↔ B)

Directional hypothesis (1-tailed)

The numbers of digits recalled will be significantly higher for those that will be read out loud than those that will be read sub-vocally. (A → B)

1 tailed direction keywords

• increase/decrease

• higher/lower

• better/worse

Directional hypotheses sentence former

The ____ will score a significantly high number of words recalled then _____.

Null hypothesis

There will be no significant difference between the number of digits recalled from those that will be read out loud and those that will be read sub-vocally (A ≠ B)

Null hypotheses sentence former

There will be no significant difference between #1 or and #2 the number of words recalled.

Experimental design

The way in which the participants are organized into groups

Characteristics of independent measures design

• Testing separate groups of people

• Each group is tested in a different condition

Pros + Cons of independent measures design

Pros	Cons
Avoids order or practice effects as they only take part in 1 condition Avoids researcher bias Avoids demand characteristics → participants able to guess aim of study leading to alterations of findings/responses Random allocation → ensures each participant has an equal chance of being assigned to a group	Using different groups of participants → participant variables → known as a type of extraneous variables (individual differences) become an issue → consequently questioning the reliability of results

Characteristics of repeated measures design

• Testing the same group of people in different conditions → same people are used repeatedly

Pros + Cons of repeated measures design

Pros	Cons	Control
Participants' variables are eliminated as each participant acts as their own control (no differences) + participants take part in all conditions	Order effect may occur → e.g. fatigue, boredom, practice effects (can get better) → greater chance of demand characteristics occurring as participants are in both conditions	Counterbalance (comparing the data between conditions → only for repeated measures) → comparison between the order effect of conditions, → try to eliminate order effect

Order effect

matters which order conditions are in

Counterbalancing

• To avoid order effects → sample groups split in half → 1 half does C#1 then C#2 → oher half does C#2, then C#1

• Eliminate order effects. Ensures changes in order of behaviors/responses are due to IV + not the sequences conditions are done

• Control confounding variables caused by fatigue, practice/boredom

• ↑ internal validity → more likely that any observed effect is generally due to experimental manipulations

Characteristics of matched pair designs measures design

• Testing separate groups of people, each member of one group could be the same age, sex, or social background as a member of the other group

• People have pre-existing characteristics

• This design can help overcome issues with repeated

Pros + Cons of matched pair designs measures design

Pros	Cons	Control
No order or practice effects occur → participants take part in 1 condition Attempts to control participant variables as they are matched	Difficult to match participants on all characteristic that could affect results → very time consuming to do so	Keeping the experiment simple: matching easier to implement when one1 matching variable + 2 groups

Single blind

When the participants last researcher doesn't know if they are in the treatment group or the control group

Double blind

When neither the participants nor the researcher know which group is which.

Correlational studies

• have no manipulated variable

• do not seek to establish casual relationships

• they have 2 or more measured variables knows as co-variables

Correlation

measures how strong the relationship is between 2 or more variables

Positive correlation

Negative correlation

No correlation

Correlational designs

participants provide data for both variables

Negative correlation coefficient

Positive correlation coefficient

Population

entire group of individuals that researchers are interested in studying → e.g. all IB students

Target population

Total group of individuals that researchers are interested in studying → e.g. SJI IB studens

Sample

smaller, manageable group of participants selected from a larger target population to take part in a research study → e.g. 1 BM, 1 eng, 1 geo class

Participants

individuals who voluntarily take part in research, providing data for studies → e.g. you

Sampling

referring to how you get participants

Probability sample

• greater chance that the sample is representative of the target population

• Results can be generalized to the wider population

Probability sample examples

• Stratified

• Random

Non-probability sampling

• Method where a researcher doesn't pick participants randomly → they use a specific criteria → e.g. based on convenience or their own judgment

• In qualitative techniques will be selected due to narrower criteria → generalizing isn't main focus

Non-probability sampling examples

• Convenience

• Volunteer

• Snowball

• Purposive

Target population

Group of people that researchers want to investigate → it can be huge → e.g. everyone on the planet or it can be restricted from the target population → the researcher then selects a group to be in their research → that group is a sample

Representational generalisability (how representative are your participants)

We can only generalize the population from which sample is drawn

Not correct to say a limitation of the study is that it cannot generalize these findings to the general population

If sample size is sufficient, you can infer that the participants characteristics are equally dispersed + fairly represented in the sample

Characteristics of random sampling technique

• Every member of TP has an equal chance of being chosen → involves identifying everyone in TP →selecting the number of participants in an objective way → e.g. pulling names from a hat

Strengths of random sampling technique

• Avoids researcher bias → researchers don’t control who is chosen as sample

• More ethical method → everyone in TP has an equal chance of being picked → especially if the research is beneficial → e.g. learning aids

Limitations of random sampling technique

• Time consuming as TP have to be identified → then have to use a generator to pick randomly

Characteristics of convenience/opportunity sampling technique

• Taking the sample from people who are available at the time + place of the study within close proximity → e.g. canteen at school

Strengths of convenience/opportunity sampling technique

• Convenient + quick for researchers as Ps are in the location of the study → more cost effective method as no time spent looking for Ps

Limitations of convenience/opportunity sampling technique

• P’s may be a group of similar people → e.g. social class → makes it unrepresentative sample of TP

• Affected by researcher bias as they get to choose their Ps

Characteristics of stratified sampling technique

• Classifying TP into categories → then choosing a sample of Ps from each category in the same proportions as they are in the population

Strengths of stratified sampling technique

• All Ps variables are taken into account in sampling → much more likely to be a representative sample of TP

Limitations of stratified sampling technique

• Difficult to identify the sub-groups in TP → more time consuming + costly to researcher

Characteristics of volunteer/self-selected sampling technique

• Individuals choose to participate in the study

• This is when the study is advertised through posters, media, etc

Strengths of volunteer/self-selected sampling technique

• As Ps volunteer → cost-effective as the researchers don’t spend anytime looking + producing a sample

Limitations of volunteer/self-selected sampling technique

• Sample is not representative of TP as they have put themselves forward → probably much more eager to please researchers → e.g. expectancy effect + social desirability

Characteristics of snowball sampling technique

• Small no. of Ps are invited + asked to invite other people they know to be part of study → due to the interest in the purpose of the study

• Usually used to obtain samples of Ps that may be difficult to reach → e.g. youth gang members, drug users

Strengths of snowball sampling technique

• Obtain samples from hard to reach Ps

Limitations of snowball sampling technique

• May create a biased sample → may not be representative of TP → making generalizing of results difficult

• Affected by researcher bias as they invite Ps who in turn pick people they know

Characteristics of purposive sampling technique

• main characteristics of Ps are defined in advance of the study

• Researchers recruit Ps that fit these conditions → proportions + sample size are not defined

Strengths of purposive sampling technique

• More representative sample of TP → Ps are chosen due to characteristics important from TP → although it said to be more representative of TP

Limitations of purposive sampling technique

• Researcher bias → due to them picking the criteria → Ps for the study