Research method

Research process

Research process

research question and hypothesis

define variables and measurements

study design

sampling

data collection

analyze data

develop conclusions and report findings

think critically and refine hypothesis

Psychology is a(n) ________________ discipline.

evidence-based

Construct

intangible, abstract attributes that are theorized to underlie observable behavior

indirectly observable or measurable

hypothetical psychological mechanism that we can’t easily observed

Happiness

subjective

abstract attribute

example of a construct

How do we measure happiness?

measure smiling behavior, which is an observable behavior that we theorize indication of the presence of happiness

it isn’t happiness itself

but we can use it to indirectly measure happiness

many ways to operationalized

Operationalizing the construct

indirectly measure construct through observation, measurement of behavior or experiences that we theorize that they are associated with it

when we define and measure a construct, indirectly through an observable behavior, then that measurement procedure becomes our operational definition of the construct

operational definition doesn’t actually refer to the construct itself, it is how we observe and define the construct through indirect measurement

Intelligence

construct

physically intangible

performance on IQ test (intelligence quotient) is an operational definition of intelligence

Research question

the broad question, idea, or problem that you wish to investigate during a research project

psychologist choose research question in their area of interest

What are the 3 types of research question?

association/relationship

difference

prediction

What do research questions come from?

inspired by personal interest

observation (own or others)

practical problems

theories

important to do a literature search and review to inform your research question

to read about relevant articles on the same area to know what have been done before and most recent

refine board question into a testable statement that can be examined during the research process

Qualities of a good hypothesis

logical: should logically follow your literature review and have a logically argument supporting your hypothesis

testable: contains observable and measurable phenomenon

narrow and specific compared to the research question

falsifiable: research study data can either support hypothesis or not support it

predictable: predict something will happen

Variable

a characteristic or condition that can change on different value for different people

anything that can vary amongst the population

can be intangible (ex. IQ) or concrete (ex. height, weight)

operationalized by using a measure in psychological research

Independent variable (IV)

causal or influential

Dependent variable (DV)

effected or influenced by IV

Continuous variable

allows for decimal or fractional values to be obtained between points on a scale when measured

ex. distance, weight, time

Discrete variable

includes separate indivisible

only allow whole values

values that can’t meaningfully exist between points on a measurable scale

ex. number of people in your family

4 scales of measurement

nominal

ordinal

interval

ratio

determined by measure used

Nominal scale

involves scale based on category

no ordering

different values don’t indicate differences in magnitude

indicate membership to different categories

ex. gender, marital status, college major, and blood type

Ordinal scale

indicate order

doesn’t tell magnitude

ex. first place, second place, third place in a running race

Interval scale

indicate actual amount of something with equal unit of measurement separating 2 scores on a scale, but there is no real zero (zero doesn’t mean a complete absence of something)

zero is arbitrary

ex. temperature

Ratio scale

indicate an actual amount of something with equal units of measurement separating each scores on a scale, but it has a real zero point

ex. distance of 2 objects

Research design

format and structure of a research project in terms of the things participants were asked to do, the way they are grouped or not, the way data and variables are collected, how the project progresses over time

deal with research question and inform type of data analysis we end up doing

5 main types of research design

descriptive

correlational

experimental

quasi experimental

non-experimental

Descriptive research design

concern with measurement and description of the natural state of individual variables as they are experienced via certain group of people

no experimentation and nothing is manipulated

focus on observation

Correlational research design

investigate relationship between variables

no experimental manipulation

variables are observed as they naturally exist

each participants provide 2 pieces of data, one for each variable

relationship between variables would mean changes in values on one variable are systematically and predictably accompanied by changes in another variable

gather evidence for association

cannot make evidence for cause-and-effect

can have a third variable that explains causal relationship, at least to some degree

Experimental research design

determining cause-and-effect in a relationship between variables

have 2 core features: involve manipulation of the hypothesized IV in a relationship

involve of carefully controlled experimental conditions in order to increase internal validity

control→ random assignment of participants to different groups

Internal validity

degree of confidence we have in a direct relationship between IV and DV in a psychological research study

by carefully controlling experimental conditions, the likelihood of encountering confounding variables for the study findings are reduced, and internal validity is strengthened

Quasi-experimental research design

investigate cause-and-effect relationship and some attempt is done to control extraneous variables

less controlled than experimental research design in that they don’t include random assignment to experimental participant’s group

ex. female vs male

Non-experimental research design

similar to correlational design

concern of demonstrating a relationship between variables but didn’t attempt to explain cause-and-effect

involve observation of two or more groups of people and one variable

variables are observed in a natural environment without manipulation and groups can be compared

What is the goal of conducting psychological research?

when we conduct psychological research, our goal is to understand some phenomenon that is experienced by a population of interest

Population

everyone of relevance to a research study

impossible to investigate everyone in the population, only a small portion of the population will be part of the study

Sample

a group of people from the population that participate in the study

What does psychological research try to achieve?

try to make a reasonable inference about things we are really interested in

select samples that are as representative as possible

Probability sampling

describes a number of sampling techniques that can be applied when the precise details of the population is known and the probability of any particular individual being selected for a research study can be determined

ex. random sampling

Random sampling

a random selection process ensure that each individual in the population has an equal probability of being selected for a research study

Strength of random sampling

free of any form of selection bias that can confound the study results

Limitation of random sampling

cannot guarantee the sample is representative of the population

often not possible in psychological research given that population’s precise characteristics are commonly unknown

Non-probability sampling

can be vulnerable to selection bias, which can affect the degree to which the sample is representative to the population

ex. convenient sampling

Convenient sampling

comprises sample that are easy or convenient to recruit

without careful control, the convenience sampling can be bias in some ways that impair researcher’s ability to generalize findings to the population

Common method to control sampling bias in convenient sampling

use quotas

involves the identification of subgroups in the accessible population, and setting recruitment quotas to ensure that they are equally representative in the population

sometimes can be even more bias, because the quotas can be set to target a predetermined set of population’s subgroup portion

can set corresponding sampling quotas

quotas may not control all the sampling bias that can occur in the convenience sampling

importantly, researchers make the best sampling decision if possible and are transparent about those decision in their reporting. this allows findings and claims about the population can be critically evaluated

Good ethical practices

ensure the welfare and dignity of those participating in a research and to ensure that the reporting of psychological is accurate and honest

Ethics in Australia is governed by?

National health and medical research council (NAMRC)

they published recommendation in the national statement on ethical conduct in human research

Guiding principle of research

merit and integrity

justice

beneficence

respect

Merit

justify by its potential benefit to humanity, either in terms of benefits to knowledge and understanding or more tangible outcomes such as improve social welfare or wellbeing for individual

more scientific so it is easier to achieve its aim

psychological research often perform initial review of relevant literature, which predicts the likely merit of the given study and additionally be valuable in informing the approach to designing that study

Integrity

research that perform according to well-founded principle of research conduct, where the researcher have conducted and reported the research honestly and transparently, whether or not the proposed hypothesis is supported

Justice

ensure that the inclusion and exclusion of participants is fair and equitable

no unfair burden placed upon any particular participant groups, that participants are not exploited in any way, that the benefit of participation is distributed fairly, and there is fair and equitable access to the benefits of the research

Beneficence

consider risks and benefits

research should involve likely benefits to either participants or the community that would justify any risks or discomfort to participants

potential risks that is associated with participation in research are minimized and controlled

researchers are responsible for clearly explaining to participant the risks and benefits of the research

this information should be given to potential participants prior to their involvement of the study

can be simple, but it is very important

once participants acknowledge the risks and benefits, they can decide whether to consent to join the study

Respect

builds on application on other core principle in research

holding regard for the value of participants and of their culture, belief, and welfare

respect privacy and confidentiality of people participating in the research project and must empower people to make decisions about their participations and the data they contribute to research studies

How do you ensure that a research project is planned appropriately, in terms of various ethical considerations?

researchers submit a research proposal to an ethics advisor, group, or committee

provide helpful advice

ask for additional requirements to be met

alternatively approve or disprove the project

a human research committee must approve a research proposal before psychological research can commence

What does the research proposal provide?

an account to the background to the project in terms of the literature or other important contexts

a detailed account of the research methodology

a description of how potential risk will be managed

a copy of all the measures that will be used

clear explanation of what participants would be asked to do and how they would be supported

copies of all documentations that participants will be given, which usually include plain language statement, a consent form, and a debriefing statement are also reviewed

Is psychological research more qualitative or quantitative?

quantitative

focus on methods that allow us to measure psychological phenomenon as precisely and unambiguously as possible

use numerical scales of measurement

ex. Likert scale

Frequency distribution

when we organize a distribution of scores by grouping the number of individuals located at each point on the scale of measurement

easily interpreted and is most commonly presented in a form of a graph or a table

Histogram

use to graph frequency distribution

easier to understand and see the distribution

Normal distribution

bell-curve

Positive skew

is a distribution is tilted to the side

the bulk is tilted to the left

major of the score is low

long right hand tail

Negative skew

opposite of positive skew

bulk falls to the right

long left hand tail

Relative frequency

express number of observation in each category as a proportion of the total number of the observation in the variable in the form of percentage

Cumulative frequency

proportion of observation for each score category accumulates to yield 100% of the observations for the variable

Box plot

low score (25%) is presented at the bottom of the graph, high score (25%) is presented at the top, middle portion (50%) is in the middle

the lines with caps at the end are called whiskers

outliers can be out of the whiskers and be presented as a blue dot

Central tendency

a summary measure that attempts to describe a whole set of data with a single value that represents the middle or center of its distribution

what is most representative or typical for a distribution

include mean, median, mode

Mean

average

add up all the score and divide that by the number of score that there were

use with interval and ratio data

normal distribution

Median

middle score

first rank the data from lowest to highest

50th percentile in a distribution

hold up against outlier

use when data is skewed or has outliers

with ordinal scales of measurement

Mode

most common score

use with discrete data

with nominal scales of measurement

Variability

how scores are spread throughout a distribution of data

tell us how representative the mean is of its distribution

if the data is packed around the mean, variability is low, which means the mean is a good representative of the scores

What does psychological research focus on?

precision

3 quantitative measure of variability

range

interquartile range

standard deviation

Range

distance covered by the entire distribution, from smallest to largest

largest minus smallest

Interquartile range

measures distance in a distribution from scores falling in the first quartile, or 25% percentile, to the scores falling in the third quartile

use it when it is appropriate to use a median

Standard deviation

average amount that scores differ from or deviate from the mean

square root of variance

Deviation score

how far each score deviated from the mean

mean minus the raw score

minus means that the raw score is below the mean

deviation score is squared to make it positive to prevent it from cancelling each other

Sum of squares

sum of squared deviation scores in a distribution

Variance

sum of squares divided by one less than sample size

squared standard deviation

Degree of freedom

n-1

Critical thinking

active process of information consumption and scrutiny

involve thinking if there are evidence to support the claim, if the evidence was of good quality, is it likely to be bias, did the claim logically followed the evidence provided, etc.

Scientific argument

consists of a number of statements (premises) that are intended to provide support for a solution

ex. deductive argument, inductive argument

Deductive argument

start with a broad premise(s)

premise(s) are used as a support to have a specific conclusion

derive suitable hypothesis to test a theory

consider valid if the conclusion necessarily follows the premise(s)

Inductive argument

include premises that include specific observation, which are intended to provide support for a conclusion in the form of a broader inference, a generalized explanation, or theory

commonly used to interpret and explain the results that are forthcoming from psychological project

doesn’t provide absolute support for its conclusion

provide probabilistic support for the conclusion

based on the premise, the conclusion is more or less likely to be true

strong inductive argument=good support (likely)

What does theories contain?

hypothetical mechanisms and intangible elements

although cannot be seen, but we assume it exists and assume it is real, because they can explain behaviors we see

a set of statements about the mechanisms underlying a particular behavior

help organize and unify different observations of behavior and relationship with other variables

generates prediction about the behavior

Validity

the degree to which the measurement process measures the variable that it claims to measure

especially important whenever an operational definition is used to measure a hypothetical construct

Face validity

unscientific form of validity

simplest and least scientific definition of validity

concerns the superficial appearance of a measurement procedure

based on subjective judgement and is difficult to quantify

high level of face validity can create problem

ex. if the purpose of measurement is obvious, the participants can see exactly what is being measured and may adjust their answers to produce a better image of themselves

Concurrent validity

demonstrate the scores obtained from the new measurement technique are directly related to the scores obtained from another, better-established procedure for measuring the same variable

establish consistency between 2 different procedures for measuring the same variable, suggesting that the 2 measurement procedures are measuring the same thing

because one procedure is well established and accepted as being valid, we infer that the second procedure must also be valid

Predictive validity

when the measurements of a construct accurately predict behavior

Construct validity

demonstrate that measurements of a variable behave in exactly the same way as the variable itself

need to examine many past researches

Convergent validity

create 2 different methods for measuring the same construct, and then showing that the 2 methods produce strongly related scores

demonstrate that different measurement procedures converge

Divergent validity

demonstrate that we are measuring one specific construct and not combining 2 different constructs in the same measurement process

differentiate between 2 conceptually distinct construct by measuring both constructs and then showing that there is little or no relationship between the 2 measurements

Reliability

produces identical results when it is used repeatedly to measure the same individual under the same conditions

stability or consistency of the measurement

Most common sources of error that leads to inconsistency

observer error

environmental changes

participant changes

Test-retest reliability

successive measurements

compare the scores obtained from 2 successive measurements

use exactly same measurement procedure for the same group of individuals at 2 different times

Parallel-forms reliability

successive measurements

different versions of the instruments are used for the test and the retest

determined by computing a correlation to measure the consistency of the relationship between the 2 sets of scores

Inter-rater reliability

simultaneous measurement

degree of agreement between 2 observers who simultaneously record measurements of the behaviors

measured by computing the correlation between the scores from the 2 observers

Split-half reliability

internal consistency

splitting the items on a questionnaire or test in half, computing a separate score for each half, and then calculating the degree of consistency between the 2 scores for a group of participants

Demand characteristics

any potential cues or features of a study that suggest to the participants what the purpose and hypothesis is and influence the participants to respond or behave in a certain way

Reactivity

participants modify their natural behavior in response to the fact that they are participating in a research study or the knowledge that they are being measured

Method of tenacity

holding on to ideas and beliefs simply because they have been accepted as facts for a long time or because of superstition

Method of intuition

information is accepted as true because it feels right

Method of authority

relies on information or answer from an expert in the subject area

Method of faith

variant of method of authority, in which people have unquestioning trust in the authority figure and accept information from the authority without doubt or challenge

ex. kids trust adults