key science skills

aim

outlines purpose of the experiment

to investigate the effect IV has on the DV

the aim is to investigate the effect of eating carrots (IV) has on eyesight (DV)

independent variable

variable that is manipulated

dependent variable

variable that is measured

population

entire group of research interest

hypothesis

prediction of experiment

it is hypoethesised that people who [are exposed to the IV] will have [prediction (increase/decrease)] on [measurement on DV] compared to [those who are not exposed to the IV]

it is hypothesied that people who eat carrots will have better eyesight compared to those who do not eat carrots

controlled variables

variables that are held constant to ensure changes in the DV are solely due to the changes in the IV

eyesight of individuals are the same level

sample

subset of populaton selected for study

random sampling

each person in the population has an equal chance of being selected for sample

number generator

only representative when picking out a lot of members from the population

time consuming

stratified sampling

population is divided into groups (age) and randomly selected in the proportion they appear in the population

most representative

more data is needed from participants

non-random sampling

selecting readily avaliable members of the population

first 10 people at the canteen

fast, easy

least representative

within subject design

same participants are used in both experimental and control conditions (with & without IV)

do people throw better with eyes open or eyes closed

→ individuals will throw with eyes open AND eyes closed

smaller number of participants required, no participant variability

order effects, more time consuming, higher rate of withdrawl rates

between subjects

individuals only do one experimental condition (either with or without IV)

do people throw better with eyes open or eyes closed?

→ individuals will either throw with eyes open OR closed

time efficient, no order effects, lower participant withdrawl rates

large number of participants required, participant variability

mixed subjects

participants are allocated to an experimental group and measured before and after exposure to IV

do people feel happier when they rest, walk or talk for 1 minute.

→ individuals will undertake one condition (rest/walk/talk) and will be measured before and after they conduct action

allows comparison of result

costly and time consuming

controlled experiement

• IV & DV present

• variables are manipulated

• time consuming

case study

in-depth investigation of a specific activity/behaviour that includes complexities that would be encountered in the real world

obtains detailed information

can’t generalise result, unable to repeat

interview

researcher asks series of questions

able to obtain rich qualitative and quantitative data

individuals may lie or not understand nature of questions

questionaires

digital or paper set questions

able to obtain rich data, anonymous

individuals may lie or not understand nature of questions

correlational studies

investigate relationship that exists between variables without manipulation

no manipulation of variables can be used to predict trends

correlation is not causation, lots of extraneous variables

extraneous variables

variables that MAY have an unwanted influence on the DV

individuals may have varying eyesight capacities

confounding variables

variables that HAS directly caused an unwanted impact on the DV

• if extraneous variables aren’t controlled, they become confounding variables.

primary data

information collected directly from the source by the researcher

useful as data is dedicated towards an investigation

costly and time consuming

secondary data

information collected at an earlier time by someone else for another purpose

cost and time effective

validity of original study is unknown

subjective data

information based on personal opinions and judgements

obtains personal opinions behind data

different people have different interpretations

objective data

information that is free from personal bias

free from personal bias

less insights behind individual thoughts behind data

qualitative data

information that is in the form of description and words

• describing quality

rich in detail and can be used to further explain data

difficult to summarise

quantitative data

information that is expressed numerically

• describing quantity

summarised easily

doesn’t provide reasoning behind data

internal validity

inside the study.

• did the study measure what is claimed to measure?

  → extraneous variables

external validity

beyond the study.

• can the results be applied to others?

  → sampling

EC: non-maleficence

no harm principle

EC: respect

considerations of participants

EC: beneficence

maximising benefits and minimising risks

EC: integrity

honest reporting of sources and results

EC: justice

moral obligation to ensure fair considerations and treatment

EG: voluntary participation

participants freely chose to be involved in the experiment

EG: debriefing

participants leave understanding true nature of experiment and wellbeing is addressed

EG: deception

participants should not be deceived unless it impacts the results

EG: confidentiality

privacy of participants personal information and anonymity of results

EG: withdrawal rights

participants can discontinue the experiment at any time

EG: informed consent

participants should be informed about nature and purpose of experiment as well as risks

control group

group who does not have exposure to IV

experimental group

group who has exposure to IV

operationalising variables

describing the detail about each variable and how it is administered and measured as much as possible

ie. eating carrots → eating 6 average sized carrots daily

conclusion

• address aim

• only specify research sample only, not wider population

• state the extent to which the researcher has answered the question

• cannot draw conclusion is poor validity

robust findings

when results remain valid across a variety of conditions

uncertainty

when something is not accurately or precisely known

reproductibility

closeness of the agreement between measurements if the same experiment was conducted at a different time under different conditions

ie. getting students to measure how quickly they can react to a stimulus, do it in the morning, and at night.

repeatability

closeness of the agreement between measurements if same experiment was conducted at the same time under the same condition

ie. getting students to measure how quickly they can react to a certain stimulus, doing it 5 times

personal errors

mistakes made by individuals conducting the study, typically due to subjective biases, misjudgments, or lack of attention to detail (not to do with measuring)

• affects precision and accuracy

  → can be avoided by repeating experiment

ie. not starting a stopwatch

random errors

unpredictable variations in the measuring process

• low precision

ie. wind impacting weighing

systematic errors

cause the reading to differ from the true value by a consistent amount each time a measurement is made

• low accuracy

• usually to do with instrument

ie. a scale is calibrated wrong

precision

how close a set of measurement values are to each other

true value

value that would be obtained if the quantity could be measured perfectly

accuracy

how close a measurement is to the true value

standard deviation

describes spread of data around mean

product, process and system development

a product, process and system designed to meet a human need

can contribute new knowledge/skills

can be time consuming and costly

modelling/simulation studies

physical or conceptual model is constructed to simulate a system and aims to imitate real life situations

• can be used to safely study new therapies

• conduct multiple trials in a short time

• investigate things that would be unethical/impossible to test in real life

• predict future events

• simulations aren’t the real thing, expensive

literature review

involves collating and analysing secondary data findings & viewpoints from multiple published sources

identify gaps in current understanding and areas for future research

only secondary data is acquired, selection bias

observational studies

researcher watches & records people’s behaviour without intervention or manipulating IV

researcher can watch without manipulation

can’t determine cause of behaviour

focus groups

facilitator conducting group discussions on a specific topic

• participants may go along with the group

fieldwork

any research involving observation and interacton with people and environment in real-world/naturalistic serttings

ie. observational studies, interviews, questionaires, focus groups, yarning circles

information on sensitive topics can be obtained

time consuming, observed behaviours subjective

yarning circles

traditional approach in aboriginal and torres strait islander culture involving non-judgemental, deep reflective discussions to exchange ideas

generalisations

can only be made if:

• results are statistically significant

• sample is representative (method of sampling is appropriate)

• whether extraneous & confounding variables have been controlled

• no known sources of error or bias

participant variability

characteristics of participants that may affect the results

ie. age

solution:

• increase sample size

• randomly allocate between subjects

• use within subjects

experimenter effect

expectations of the researcher affect results of an experiment

• biased interactions

• biased data collection

• confirmation bias

solution:

• double blind procedure → experimenters and participants are unaware of particular experimental conditions

demand characteristics/placebo effect

participant believes they are receiving experimental treatment and they respond in accordance to their belief

solution:

• deception

• single blind procedure → participants are unaware of experimental conditions

situational variables

ie. different weather, timing

solutions:

• controlled variables

• standardised instructions & procedures → ensuring each participant in an experiment receives the exact same instructions

advantages of mean

• makes use of all data values

disadvantages of mean

• misleaded by extreme values

advantages of median

• not affected by extreme values

disadvantages of median

• does not take into account all values

advantages of mode

• useful when data are in categories

disadvantages of mode

• not useful when there are several modes

classification and identification

classification - in research, involving arranging objects/events/phenomenon

identification - recognising phenomenon s belonging to a particular set

• help people determine if their behaviour is usual or unusual

• allow for efficient processing of large amounts of information

• labelling through identification leads to stereotypes/prejudice

occupational health and safety (OHS)

the issues, health, safety & welfare that must be protected

• risk assessment

risk assessment

process that considers identifies, and reduces physical and psychological risk

outliers

value that lies long way away from other values