Slides in this set
Hypotheses & the IV + DV
Hypothesis - It's a precise and testable statement of the relationship between two variables.
Research hypothesis: proposed at beginning of piece of research + often generated from a theory.
Null hypothesis: What you assume is true during the study. Any data you collect will either back this
assumption up or it won't. If the data doesn't support you null hypothesis, you reject it and go with
your alternative hypothesis instead. Very often the null hypothesis is a prediction that there will be no
relationship between variable in a study any correlation is due to chance.
Experimental (alternative) hypothesis: If the data forces you to reject your null hypothesis, then you
accept your experimental (alternative) hypothesis instead. So if your null hypothesis was that two
variables were not linked then your alternative hypothesis should be that they are. Or you can be more
specific saying why they are linked using directional hypothesis.
Directional hypothesis: States the direction of results.
Non directional hypothesis: Predicts simply that there will be a difference between two conditions or
two groups of participants, without stating the direction of the difference.
Eg/ `People remember more when they study in short bursts'
IV variable changed by experimenter = `study in short bursts'
DV effects of IV on another variable are observed or measured = `remember more'
The IV needs to have levels, so there is something to compare it to, so `people remember more when
they study in short bursts rather than over longer periods'. Then needs to be operationalised, which
involves defining `short bursts' (3 x 10 minutes), `longer periods' (1 hour) and 'remember more'
(something to learn and a test afterwards). The operationalised hypothesis would be: People get more
questions right on a test of recall when they study in short bursts (3x10 minutes) then when studying
for longer periods (1hour).…read more
Lab experiments: Experiment conducted in special environment where variables can be controlled.
PP's are aware they're taking part, though may not know the true aims.
Field experiments: Experiment conducted in more natural environment (anywhere outside a lab). IV
is still deliberately manipulated by the research, but pp's often not aware of participating.
Natural experiments: Experiment in which the researcher can't manipulate IV directly but where it
varies naturally and the effect can be observed by the DV .
Method Advantages Disadvantages
Lab Replication strict controls mean that you Lack Ecological Validity might not measure "real
can run the study again to check the findings life" behaviour
Causal relationships ideally it's possible Ethics deception is often used, so hard to get
to establish cause and effect. informed consent without messing results up.
Field Ecological validity these are less artificial Less control confounding variables are more likely
than lab ones so relate to real life better in a natural experiment Ethics pp's who didn't agree
Demand characteristics these can be to take part might experience distress + can't be
avoided if the participant doesn't know that debriefed. Observations must respect privacy.
they are in a study.
Natural Ethics make it possible to study variables Ethics deception often used making informed consent
that it would be unethical to manipulate e.g. the difficult. Also confidentiality may be compromised if the
effects of smoking during pregnancy on birth community is identifiable
weight Casual Conclusions cannot be drawn because there
is no deliberate manipulation of the IV.…read more
This is a set of procedures used to control factors such as pp variables. There are 3 types:
1) Repeated Measures Design: Each pp takes part in every condition under test
Strengths Weaknesses Solution
Less pp's needed to quicker One condition may be harder than Make equivalent tests to
and cheaper. another (extraneous variable) and make both conditions
Participant variables shouldn't would affect accuracy of results. equal.
exist because same people do On 2 test, pp's may have guessed Produce cover story about
each condition. experimental aims and may influence test purpose (single blind)
answers. Use counterbalancing.
Order of conditions may affect Either AB/BA or ABBA.
performance (order effect). Eg/ the
practice affect (doing it twice).
2) Independent Groups Design: Pp's are allocated 2+ groups representing different
experimental conditions. Allocation is usually done using random techniques.
Strengths Weaknesses Solution
No order effects no one gets No control over pp variables. Randomly allocate pp's to
better through practice (learning conditions.
effect) or gets worse through Need twice as many pp's. Be prepared to spend money.
being bored or tired (fatigue
3) Matched Pairs Design: Pairs of pp's are matched in terms of key variables such as age or
IQ. One member is placed in the experimental group and the other in the control group.
Strengths Weaknesses Solution
No order effects because pp's Very time consuming to match Restrict matching variables to
are only completing one pp's and probably have to start make it easier.
condition. with large group which can be
Lowers participant variables expensive.
(although still no full control). May not control all pp variables Conduct pilot study to
Less pp's needed so quicker because you can only match on consider key variables.
and cheaper. variables known to be relevant.
Counterbalancing - This is basically mixing up the order of the tasks. This can solve order
effects in repeated measures designs.
Way 1) AB or BA: Divide pp's into two groups.
Group 1 Does condition A then B.
Group 2 Does condition B then A.
Way 2) ABBA: All pp's take part in all 4 conditions, twice.
Trial 1 Condition A (Morning)
Trial 2 Condition B (Afternoon) Compare scores on
Trial 3 Condition B (Afternoon) trials 1+4 with 2+3.
Trial 4 Condition A (Morning)…read more
Observations (non-experimental method):
Naturalistic observation: Observation carried out in natural setting, in which
investigator doesn't interfere, only observes behaviour, although likely to involve
structured observation (using categories to organise observations).
Controlled observation: Behaviour is observed under controlled conditions and PP's are
likely to know they're part of an experiment.
Observational techniques are used within the above types to ensure it's rigorous and
Structured Observations: Research uses `systems' to organise observations.
1) Behavioural categories - How to record behaviour you're interested in. Researcher
needs to break up stream of behaviour into different categories (operationalisation).
The categories should be objective (record explicit actions), cover all possible
component behaviours and be mutually exclusive (not marking 2 categories at once).
2) Sampling procedures - Continuous observation can be difficult because there would
be too much data to record, so there are 2 types of systematic sampling methods:
Event sampling Counting number of times event occurs in target individual.
Time sampling - Recording event in given time frame.
3) Unstructured Observations: Researcher records all relevant behaviours with no
system. However this can mean there is too much to record and behaviours recorded
are those most visible but not most important.…read more