Key Science skills Psych

Controlled experiment

investigates the relationship between one or more IV(s) and a DV, while all other variables are controlled.

Controlled experiment strengths and limitations

strengths:

• can idenitify a cause and effect relationship between IV(s) and DV

• can be repeated to gather more data and increase reproducibility and repeatability of results

Limitations

• strictly controlled conditions/variable can be hard to maintain, so results can be influenced by extraneous variables

• participants behaviours may be influenced by the artificial nature of the setting

• external validity can be low due to artificial nature

Between subjects design

investigation design in which participants are allocated to only one condition of the experiment

Between subjects design strengths and limitations

strengths:

• time efficient, as both groups can be tested at the same time

• lower rate of participant withdrawal

Limitations:

• more participants needed

• less control over extraneous variables of participants

Within subjects design

All participants in the sameple completing all conditions of the experiment

Within subjects design strengths and limitations

strengths:

• no extraneous variables

• fewer participants needed than in a between subjects design

Limitations

• more time consuming, as both cannot be tested at the same time

• higher rate of participant withdrawel, as DV has to be measure multiple times

Stratified sampling

When the population is divided into groups based on relevant characteristics, participants are selected from each group in proportion to how they appear in the population

• time consuming and expensive

• may be unethical or impossible to obtain such groups

• if sample is big enough it could improve external validity

random sampling

When every member of a population has an equal chance of being selected for the sample

• reduces bias → more likely to represent population

• time consuming

Neither stratified nor random sample (convenience)

Sample selected using the quickest and easiest means possible, selecting readily available people

• However not representative of the population

Random error

Affect the precision of a measurement by creating unpredictable variations in the measurement process, resulting in a spread of readings.

• Maybe be produced by a limitation in the instrument, an environment factor or a slight variation in the procedure

Systematic error

affect the accuracy of a measurement by causing readings to differ from the true value by a consistent amount or by the same proportion each time a measurement is made

• All the readings are shifted in one direction from the true value

• May be produced by observational error, imperfect instrument calibration or environmental interference

Personal errors

mistakes, miscalculations an dobserver errors made when conducting research

• to remove them, the investigation should be repeated correctly, and errors should NOT be included in the reporting and analysis of the data

uncertainty

• lack of exact knowledge of the value being measured

• NOT the same as error - results cna be uncertain even when measurements are taken correectly.

Repeatability

Repeatability refers to the degree to which a specific research investigation obtains similar results when it is conducted again under the same conditions on all occasions.

• Can the same person get the same results using the same method under the same conditions?

• Focuses on consistency within the same setup.

→ Think: If I re-do the same experiment, do I get the same result?

Reproducibility

Reproducibility refers to how close the results are to each other when an investigation is replicated under changed conditions.

• Are the results consistently accurate across different times, people, or tools?

• Focuses on overall consistency and trustworthiness of the measurement.

→ Think: Is this method giving me the right result every time, no matter who uses it?

• Is my sample representative of the population?