Hypotheses, Variables and Sampling
The Starting Point of Scientific Research
Every psychological study starts with a question. A researcher notices something about human behaviour and wants to know why it happens, or whether one thing influences another. But noticing something and testing it scientifically are very different tasks. Before any data can be collected, a researcher must be precise: they must state exactly what they expect to find, define what they are measuring and ensure they have an appropriate group of participants. These steps are the foundation of scientific psychology.
Hypotheses
A hypothesis is a precise, testable prediction. It states in advance what the researcher expects to find. A hypothesis must be operationalised — meaning both the independent variable and the dependent variable must be defined in terms of how they will be measured, not described vaguely.
Compare these two predictions: “People who revise more will do better” versus “Participants who revise using flashcards for five hours will recall more words from a 20-word list than participants who reread their notes for five hours.” The second is a well-operationalised hypothesis because both conditions of the independent variable (flashcards vs rereading) and the dependent variable (words recalled from a 20-word list) are clearly defined and measurable.
A well-operationalised hypothesis states both conditions of the independent variable AND a clearly measurable dependent variable. Without operationalisation, a hypothesis cannot be properly tested, and both elements are needed for full marks.
Null and Alternative Hypotheses
Every study involves two complementary hypotheses. The null hypothesis states that there is no difference or no relationship between the variables — any result found is due to chance. For example: “There will be no significant difference in word recall between participants who revise with flashcards and those who reread notes.”
The alternative hypothesis (also called the experimental hypothesis) states that there will be a difference or relationship — this is what the researcher expects to find. For example: “Participants who revise using flashcards will recall significantly more words from a 20-word list than those who reread their notes.”
Statistical testing after the study will determine whether the data allows the null hypothesis to be rejected. If results are ‘significant’ (unlikely to be due to chance), the null hypothesis is rejected and the alternative hypothesis is supported. Psychologists do not ‘prove’ their hypotheses; they provide evidence for or against them.
A independent variable (IV) is the one the researcher changes or manipulates. In the flashcard example, the IV is the revision method (flashcards vs rereading).
A dependent variable (DV) is what the researcher measures — the outcome. In the flashcard example, the DV is the number of words recalled from the 20-word list.
Extraneous variables are any other variables that could affect the DV that the researcher has not controlled. For example, if one group of participants had a higher baseline memory ability, that could explain any difference in word recall rather than the revision method. Controlling extraneous variables is essential to ensure results reflect the IV rather than something else.
Target Populations and Samples
A researcher almost never studies every member of a group they are interested in. Instead, they select a sample from the target population. The sample should be representative — it should reflect the characteristics of the target population as closely as possible so that conclusions can be generalised beyond the study itself.
Random sampling gives every member of the target population an equal chance of being selected. The researcher lists all members of the population and uses a random process (lottery, random number generator) to select participants. Strength: unbiased and representative. Weakness: requires a complete list of the population, which is often difficult to obtain, and non-responders may create a biased final sample.
Opportunity sampling selects whoever is available and willing to take part. A researcher standing at a university entrance and asking passers-by is using opportunity sampling. Strength: quick, easy and inexpensive. Weakness: biased towards people who are easily available at that time and place, so the sample may not be representative of the target population.
Systematic sampling selects every nth person from a list. For example, if there are 1,000 people in the population and you need 100 participants, you would select every 10th person. Strength: straightforward to carry out and reduces researcher bias. Weakness: may still be biased if the list has a pattern that makes every nth person systematically different.
Stratified sampling divides the population into subgroups (strata) based on relevant characteristics (e.g. age, gender) and then randomly selects participants from each stratum in proportion to their size in the full population. Strength: ensures the sample reflects the population structure, making it highly representative. Weakness: time-consuming and requires detailed knowledge of the population’s structure.
Four sampling methods: random (equal chance, unbiased but hard to obtain full list), opportunity (quick but biased towards available people), systematic (every nth person, straightforward but potential pattern bias), stratified (proportional subgroups, highly representative but time-consuming).