Experiments Flashcards
What are the three types of experiment?
- Laboratory
- Field
- Quasi (Natural)
What is a laboratory experiment?
The IV is manipulated by the researcher and the experiment is carried out in a lab or other contrived setting (controlled)
What is a field experiment?T
The IV is manipulated by the researcher but the experiment is carried out using the participants normal surroundings
What’s a Quasi experiment?
A natural experiment where the IV occurs naturally and cannot by manipulated by the researcher (eg. sun)
Advantages of laboratory experiment
- Tighter control of variables
- Easier to comment on cause and effect as setting is controlled
- Relatively easy to replicate
Disadvantages of laboratory experiment?
- Demand characteristics - participants are aware of experiment so may change/alter their behaviour
- Artificial environment - has low ecological validity so its difficult to generalise to other situations
- Experimenter effects - bias when expectations affect behaviour
Advantages of field experiment
- Less artificial as its set in the real world situations (more true to life - ecological validity)
- Valid as people are unaware of experimental conditions and are in their natural environment so act genuinely
Disadvantages of field experiment
- Less control over extraneous variables
- Ethical problems; don’t usually gain informed consent of those involved or they’d give away the aim of the study
Advantages of quasi experiments?
- Useful when its unethical to manipulate the IV
- Studies the ‘real effect’ so there is increase realism and ecological validity
Disadvantages of quasi experiments/
- Difficult to get participants
- Expensive/Costly and takes time because you have to wait for IV to occur
- Confounding experiment variables are more likely so less reliable
What are the three experimental designs?
Repeated measures design
Independent measures design
matched participant design
What is a repeated measures design?
involves the use of the same people in each experimental condition
What is an independent measures design ?
Involves using different people in each condition
What is a matched participant design?
Involves the use of different people in each experimental condition but an attempt is made to make the participants as similar as possible on certain key variables
What is a laboratory experiment?
The IV is manipulated by the experimental and is carried out in a lab or contrived/controlled setting
What is a field experiment?
The IV is manipulated by the researcher but it is carried out in the participants normal surroundings/environment
What is a Quasi experiment?
A natural experiment where the IV is naturally occurring (not manipulated by the researcher
What are the advantages of a laboratory experiment?
- Tighter control of variables
- Easier to comment on cause and effect relationship
- Relatively easy to replicate
What are the disadvantages of a laboratory experiment?
- Demand characteristics - participants are aware of experiment so may change behaviour
- Artificial environment- low ecological validity so difficult to generalise to other situations
- Experimenter effects - bias when expectation affects behaviour
What are the advantages of a field experiment?
- Less artificial - set in the real world situations so it’s more true to life (valid)
- Valid as people are unaware of experimental conditions s and they’re in their usual environment
What are the disadvantages of a field experiment?
- Less control over extraneous variables
- Ethical problems - don’t usually gain informed consent of those involved or they’d give away the aim of the study
What are the advantages of a Quasi experiment?
- Useful when unethical to manipulate the IV
- Studies the ‘real effect’ so there is increased realism and ecological validity
What are the disadvantages of a Quasi experiment?
- Difficult to get participants
- Expensive and requires lots of time and effort because you have to wait for the IV to occur
- Confounding experiment variables are more likely so results are less reliable
What is a repeated measures design?
This involves using the same people in each condition
What is the independent measures design?
This involves using different people in each condition
What is a matched participant design?
This involves using different people in each condition but an attempt is made to make the participants as similar as possible on certain key characteristics
Done by testing individual in key variables and pairing them based on similar scores
What are the advantages of repeated measures design?
- Requires less participants saving time and money
- Participant variables were controlled and will not affect results as the same people are used to
- Half the no. of participants are needed than the other two designs
What are the disadvantages of the repeated measures design?
- Order effects : getting worse due to fatigue/getting better due practice, Can affect the results of the second condition, to reduce order effects have ti create additional test materials with identical difficulty level (hard to achieve)
- Demand Characteristics- participants want to produce the correct reults
What are the advantages of independent measures designs?
- Avoids effects (only take part in one condition)
- Demand characteristics are less likely (can’t guess aim of study as only experienced one condition)
- Can use same task in each condition, controlling for any differences in this as a possible extraneous variables
What are the disadvantages of an independent measures design?
- Participant variables can affect behaviour
- Bigger sample is needed - uses more time and money (twice as many needed for same no. of results as repeated measures design
- Results can be affected by participant variables (inevitable individual differences)
What are the advantages of a matched participant design?
- Avoids order effects (one participant per condition)
- Avoid participant variables
- You get benefits of both other designs: order effects are avoided as you only take part in one condition and participant variables aren’t a problem
What are the disadvantages of a matched participant design.
- Requires a bigger sample which requires time and money
- It’s not practical- almost identical participants is difficult to get and factors like tiredness and motivation affect behaviours
- More complicated as participants need pre-testing on relevant characteristics so matched according
What can extraneous variables become if not controlled?
Confounding variable - obscure effects of IV
What are the two types of extraneous variables ?
Participants variables
Situational variables
How can participant variables be controlled?
Either have same/very similar people on each condition (by using repeated measures/matched participant design)
If using an independent measures design make a point of ALLOCATING PARTICIPANTS TO CONDTIONS ON A RANDOM BASIS so that participant variables are more likely to be evenly spread
How to control situational variables like order effects?
Having different people in each condition will avoid this problem (use independent/matched participant design)
If using a repeated measures design then this should be counter-balanced
- participants are split into two groups which group 1 doing condition A then B and group 2 doing condition B then A
How to control situational (environmental factors) variables?
Impose CONTROLS on the experiment to ensure there are as few differences as possible
How to control situational (demand characteristics ) variables
Do not tell participants the aim of the investigation (single blind)
What is a double blind trial?
When neither the participant or the researcher know the aim of the study
A double blind procedure eliminates the danger of (bias)?
Researcher bias
What is researcher bias?
When the researcher allows their hopes and expectations for what data should look like to affect the data they choose to hold onto
What are the two types of hypothesis?
Alternative (One or Two Tailed)
Null
What does an alternative hypothesis do?
This predicts how the IV WILL affect the DV
What are the two types of alternative hypothesis?
One-Tailed (Directional)
Two-Tailed (Non-directional)
What does a null hypothesis do?
It predicts that the IV will NOT have an effect on the DV
Instead any difference see will be due to chance factors
What key word is used in hypotheses?
Significant
What is a two tailed hypothesis?
One which predicts the IV will have a significant effect on the DV but does not predict the direction the effect will go in
What is a one tailed hypothesis?
One which predicts that not only the IV will have a significant effect on the DV but also the direction the effect will go in
How do you start to write a two-tailed hypothesis?t
There will be A significant difference between
How do you start to write a one-tailed hypothesis?
People who… will
Give an example of a hypothesis for the (lavender oil in bath water related to sleep) study?
One- tailed:
People who have 20ml of lavender oil in their bedtime bubble baths will get to sleep more quickly (30 minutes or less) than those who have unscented lavender water added to their baths
Two-Tailed:
There will be a significant difference between the babies who have had 20ml of lavender oil added to their bedtime bath than those who have had 20ml of unscented lavender water added to their bath and the time it takes to go to sleep (30 minutes or more)
What type of hypothesis?
‘Rainy weather has a significant affect on people levels of happiness’
Two-tailed
What type of hypothesis:
‘Individuals are significantly more likely to conform when in groups of five than when in pairs’
One-Tailed
What is operationalisation?
The process of making variables physically measurable/testable
Do hypotheses need to be operationalised?
Yes 👍
How do you operationalise the following IV:
- Sporty/Unsporty
- Older/Younger
- Healthy/Unhealthy
Set up into two discrete conditions;
- Sporty individuals play 2 or more hours of individual/team sport per week (organised)
- Older is 65 + years and Younger is 30 or less
- Healthy people eat 5 or more pieces of fruit and veg per day whilst non-healthy people eat two or less
Operationalise the following DV’s;
- Reaction time
- Memory ability
- How creative people are
- Measured in the number of millimeters along the ruler at which the participant catches the ruler in the ruler drop test
- Score on memory test (in %) over 60% is good
- Measure how many minutes per week people spend drawing/painting etc./ GCSE art grades
Define what is meant by target population?
The group of people the researcher is interested in studying
Define the term sample?
The actual group of participants used in the research
Having a mixed/diverse sample allows the researcher to do what with the research?
Generalised to rest of population (High population reliability)
What’s a self-selecting sample?
When people VOLUNTEER to take part in the study
Often adverts, posters or leaflets will be distributed which contain details about the research and contact details for the participant if they wish to take part
How is a self-selecting sample collected?
Often adverts, posters or leaflets will be distributed which contain details about the research and contact details for the participant if they wish to take part
What is a opportunity sample?
A sample of participants produced by selecting those who bare the MOST READILY AVAILABLE at a given time and place selected by the researcher.
What’s a random sample?
A technique where each member of the target population has an EQUAL CHANCE of being selected
What’s a snowball sample?
When participants are asked to contact their friends or family members to ask them to also take part in the research. In turn they they ask other people to
What’s a strength of a self-selecting sample?
- Ethical because participants volunteer
- Relatively easy and participants are likely to turn up
What are the weaknesses of a self-selecting sample?
- Biased based on who volunteers
- Possible time and cost of advertising
What are the strengths of a opportunity ample?
- Quick to gather participants
- Easy as no advertising or selection process
What are the weaknesses of a opportunity sample?
- Biased based on where you go and who’s available at that time
- Less ethical as participants may feel obliged to take part
What are the strengths of a random sample?
Should be representative of the target population
What are the weaknesses of a random sample?
- Can be difficult to include all people form the target population (eg. Might be ill that day)
- Can still be biased based in who’s willing to take part
What are the strengths of a snowball sample ?
- Easy as only requires finding a few participants before recruiting the rest
- Relatively cheap as low/no advertising process
What are the weaknesses of a snowball sample?
- Bias as all participants are likely to know each other and therefore have many similarities
- May no get enough participants
What is bad about a small sample size?
It is not reliable and cannot observe a consistent effect to generalise to the rest of the target population
What’s a strength of a large sample size?
Can claim reliability (consistent effect) but on a practical level the sample may be harder to manage
What is primary data?
First hand research collected directly by the experimenter
What is secondary data?
Second hand data which has been discovered in prior research and helps to analyse information which already exists
What is quantitative and qualitative data?
They are numbers, raw scores, percentages etc.
They are qualities of things (descriptions, words, meanings) comes from asking open ended questions
Strengths of QUANTITATIVE data?
- Easy to analyse (see patterns on tables)
- Easy to summarise and present in tables and graphs
- Objective data analysis (Facts which are neutral, scientific and fair)
Weaknesses of QUANTITATIVE data?
- Less in depth/detailed and can’t tell ‘why’ behaviours occurred
- Unexpected behaviours May not be counted
What are the strengths of QUALITATIVE data?
- Often ‘richer’ and more detailed
- Unexpected behaviours can be captured and/or explained
What are the weaknesses of QUALITATIVE data?
- Difficult to analyse (see patterns)
- Difficult to present in summarised forms
- Subjective bias interpretation of data using opinions and personal beliefs (researcher bias)
What are the two ways of summarising data?
Central Tendency
Measures of dispersion
What are the three measures of central tendency?
Mean
Median
Mode
What are the three measures of dispersion?
Range
Variance
Standard Deviation
How do you calculate the mean
Add up all the scores and then divide them by the number of participants in that condition
How do you calculate the median?
Place all scores in rank and see which is the middle value
How to calculate the mode?
See which result occurs most often
What are the advantages of the mean?
Most powerful measure of central tendency
Includes all data
What are the disadvantages of the mean?
Can be distorted by extreme outlier scores
What are the strengths of the median?
Not distorted by extreme outlier scores
What are the weaknesses of the median?
May not be representative of the full set of scoers
What are the strengths of the mode?
Not influenced by extreme scores
Can be used for non-numerical data
What are the weaknesses of the mode?
Not useful if there are many modes or none at all
How to calculate the range?
Largest score minus smallest score
What is variance?
Indicates how spread apart the data is within each condition
What does it mean when the variance is smaller?
The participants scores are less spread apart from the average score
How do you calculate the variance?
To calculate the variance, calculate the mean score per condition in the experiment
For each participant then subtract the mean score from their score to get ‘d’ (difference)
Then square each ‘d’ score (d^2)
Add all the d^2 scores together and get the sum of the differences squared
Calculate the mean of the d2 scores added together by dividing the figure by n-1 (n = number of participants)
How do you calculate standard deviation?
Square root variance
When talking about standard deviation what must you do?
More Widely spread of data
Less wide spread of data
What are the advantages of range?
Can be very helpful at easily and Quickly giving a sense of how dispersed the data id
What are the disadvantages of the range?
- Doesn’t give an indication as o whether the spread of scores is evenly distributed or clustered towards the top/bottom of the data set
What are the advantages of the variance?
It is an improvement over range as it takes into account of all values of data in the sey
What are the disadvantages of the variance?
It is more complicated to calculate than the range
What are the advantages of standard deviation?
Provides an indication of the extent to which data is spread out from (or clustered around) mean scores
What are the disadvantages of standard deviation?
Time - consuming and more difficult to calculate than the range score
What is a longitudinal study method?
repeatedly collect data from the same sample over an extended period of time
What is a cross-sectional snapshot study method?
a research study that is carried out in a very short period of time.
What are the strengths of a longitudinal study?
- Participant variables are controlled for
- Large amount of quantitative and qualitative data can be collected
- Researchers can see the precise moments when children change
What are the weaknesses of a longitudinal study?
- Participant attrition - people drop out of the study
- Time consuming for all results to come in reducing replicability of study
- Expensive
- Researchers can get emotionally attached so lose objectivity
- Demand characteristics may be a problem if they work out what the researcher is looking for
- Likely done with a small sample so results can be distorted by anomalies
What are the strengths of a snapshot study?
- Results are obtained more quickly
- It is likely there will be less participant attrition
- Easier to do with a large sample so they can establish if results are reliable (consistent)
What are the weaknesses of a snapshot study?
- There will be participant variables (eg. individual differences) between participants making comparisons difficult
- You only get a snapshot of development and do not see patterns/changes in an individual over a longer period
