Psychological Investigations
AS OCR Psychology - Psychological Investigations
- Created by: Hannah
- Created on: 01-04-12 16:15
General Things You Need To Know
Hypotheses (Only in Experiment and Correlation)
1. Experimental/Alternate/Research Hypothses - predicts there will be a difference in results between the conditions.
E.g. Gender & Memory
1 Tailed - Males will perform better on a memory test than females, as measured by the number of photos correctly recalled out of 20.
2 Tailed - There will be a difference between males and females and their performance on a memeory test, as measured by the number of picture correctly recalled out of 20.
2. Null Hypotheses - predicts there will be no difference betwwen the conditions.
E.g.
There will be no difference between males and females and their performance on a memory test, as measured by the number of pictures they can recall out of 20.
General Things You Need To Know
Independent and Dependent variables
IV - the thing that you change betwwen the conditions
DV - thing that you meaure
Confounding Varables (Extraneous variables) - factors that were not controlled but that could affect the results e.g. participants mood.
General Things You Need To Know
Sampling Techniques
Oppourtunity Sampling - using whoever is avaliable at the time from the given population.
+ Not time consuming - the sample may be biased e.g. all girls
Random Sampling - every person in the given population is out into a list and their names are randomly picked. e.g. putting names in a hat.
+ everybody in the population has an equal chance of being picked
- doesn't guarantee a sample that is totally representitive of the popultion
Self - Selected (volunteer) sampling - involves participants electing themselves often through an advert (e.g. Milgram)
+ - people have gvien their consent (ethics), esay to get sample and less time consuming - most of the target population are unlikely to respond, respondents atypical, sample would probably be biased.
General Things You Need to Know
Quantitative and Qualitative Data
Quantitative Data - numbers, percentages, quantites etc..
+ easily comparable, more reliable, objective - lacks depth, less valid
Qualitative Data - meanings, understandings, words etc..
+ detailed, more valid, increaed depth, more subjective - often smaller samples, less reliable, more subjective, harder to generalise
General Things You Need To Know
Descriptive Statistics
Mean - average, add all scores together and / by how many there are
Median - middle value, if even number add two middle numbers together and / by 2
Mode - most frequently occuring value
Nominal data - data that was seprate catergories
Ordinal data - ordered in some way e.g. ask people to list their favourite football teams in preference order
Ratio data - true zero point and equal interval between two points on a scale
General Things You Need To Know
9 Ethical Issues:
1. Consent
2. Confientiality
3. Collegues - if you observe a collegue conducting unethical research, you are obliged to report them
4. Deception
5. Debrief
6. Withdrawl from investigation
7. Giving Advice - cant give psychological advice unless you qualified to
8. Protection of participants
9. Observation - you are allowed to observe people withouth their consent if its in a place where they would ordinarily be observed
Self-Report
Self-report methods are investigative techniques that focus on the participants perspective. They are more likely to produce findings that reflect one participants point of view. The main two types are questionnaires/surveys and interviews.
There are 3 types of rating scales:
1. Bi-Polar Rating scale - Involves having answers the go from one end of a scale to another.
E.g. How would you rate Disco Dancing?
Brilliant Good Okay Poor Rubbish
= = = = =
+ easy to compare data
- fixed responses: 'impose a framework on the participants reality'
Self-Report
2. Checklist Rating scale - Involves giving participants a list of things that can choose from, as to what applies to them.
E.g. What do you enjoy doing at the weekend?
seeing friends/ cinema/ swimming/ listening to music/ working
+ produces quantitative data, whcih is easy to compare
- fixed responses: 'impose a framework on the participants reality'
3. Graphical Rating scale - involves giving participants a scale.
E.g. How do you feel about testing beauty products on animals?
| | | + participants are able to more acurrately Angry Neutral Happy reflect their feelings
- difficult to compare findings
Self-Report
Open and Closed Questions
Open Questions - questions that don're require a fixed repsponse
+ participants can write what they feel, produces qualittative data
- difficult to compare
Closed Quesitons - have fixed responses
+ produces quantitative data so easy to compare
- fixed responses, 'impoe a framework on the participants reality'.
Self-Report
Strengths and Weaknesses of Self-Report
Quesstionaries and Surveys:
+ Record the participants experiences
+ Response rate can be high
+ Can sample a very large target population
+ Can contain questions on issues that are difficult to talk about
- Often Influenced by bias; difficult to design well
- Response rate can be low
- The data gathered can be complex and be difficult to interpret
- Participants may not answer answer truthfully
Self-Report
Interviews:
+ Provide large amounts of qualitative data
+ Enable the researcher to respond to the participants experiences
+ More likely to gain an insight into the participants world
+ Can use body language to confirm verbal response
- Difficult to carry out
- May be influenced by experimenter biased
- Can be difficult to analyse
- Take a long time to carry out
-Participants may not answer truthfully
Self-Report
What is meant by Reliablity?
Conducting the test more than once and getting the same results (consistent results)
How could you ensure that your questions/questionnaire/interview was reliable?
Use the same set of questions and same set of instructions that you fie to the participants and give them the all the same amount of time to complete the task.
What is meant by Validity?
When something measures what it claims to measure.
How could you ensure that your questions/questionnaire/interview was valid?
Ensuring questions are specific and focused on the topic in hand.
Observation
Participant Observation - Observer gets invovlved in the study
Non-participant - observer 'sits back' and observes behaviour
Covert - observer is 'undercover'
Overt - Participants are aware of observation
Structured - behaviour is broken up into categories and a checklist is used
Unstructured (naturalistic) - natural situation is studied
Direct observation - the observer directly observes the behaviour
Indirect observation -observations of data, such as observing TV adverts
Time sampling - observer decides on a time interval, such as once a minute, and the observer notes any behaviour and this interval.
Event sampling - observer keeps account of each time a particular behaviour is observed.
Observation
Behaviour checklist/coding system (e.g. Bandura = aggression).
- should be objective
- cover all possible component behaviours & avoid 'waste basket' category
- shouldn't have any overlapping categories
Participant Observation:
- Identify specifics - how to record the behaviour, e.g. behaviour checklist/coding system/record on camera/microphone...
- small pilot study
- ethics e.g. consent, access - how do you ask permission to 'enter the environment'?
Structured Observation:
- coding system/behaviour checklist
- time/even sampling
- who/where/when/why?
- Inter-rater reliability
Observation
Strengths of Observation:
1. A means of conducting preliminary investigations in a new area of research, to produce hypotheses for future investigations.
2. Able to capture spontaneous & unexpected behaviour
Weaknesses of Observation:
1. Observers may 'see' what they expect to see - observer bias
2. Poorly designed behaviour checklist/coding system reduces reliability (low inter-rater reliability)
Ethical issues of observation:
Informed Consent, right to withdraw, protection from harm, deception, debrief.
Observation
Reliability - how consistent any measure is.
How could you ensure the your observation was reliable?
Behaviour checklist could be used by 2 different researchers then checked for accuracy.
Validity - extent to which research has measured what it intended to measure.
How could you ensure that your observation is valid?
Using more than 1 observer & averaging data across observers to balance out any bias.
Experiment
An experiment involves the manipulation and measurement of variables. For example, in a lab experiment the researcher manipulates one variable and then measures the affect on another variable.
Possible experimental hypothesis: gender and memory
One Tailed Experimental Hypothesis - Males will perform better than females on a memory test as measured by the number of pictures correctly recalled out of 20.
Two tailed Experimental Hypothesis - There will be a difference between males and famles and their performance on a memory test, as measured by the number of pictures recalled out of 20.
Null Experimental Hypothesis - There will be no difference between males and females and their performance on a memory test as measured by the number of pictures correctly recalled out of 20.
Experiment
Possible Independent and Dependent variables in an experiment, how each could be measured and an alternative way of measuring: Gender and Memory
Independent Variable = gender
= measured by asking participants
= alternative way of measuring could be to observe participants
Dependent Variable = memory
= measured by showing 20 pictures and seeing how many they can recall
= alternative was of measuring could be showing 20 words and seeing how many they can recall
Experiment
Writing an appropriate procedure:
Aims
Hypotheses
Population
Sampling
Technique
Sample
Test
Administrate
Collate
Data
Experiment
Appropriate descriptive statistics:
Calculate the mean, median and mode for each condition, and place into tables an bar charts.
Correlation
Positive and negative correlation
Positive Correlation - as one variable increases the value of the other variable will also increase.
Negative Correlation - as one variable increases the value of the other variable will decrease.
Scattergraphs show correlation between 2 sets of data. The extent to of the correlation is described using a correlation coefficient, whcih is a number between +1 and -1.
Hypotheses:
- One tailed hypothesis - There will be a negative relationship between how people perceive the attractiveness of an animal, as measured by a Bi-Polar rating scale, and how old they are, as measured by the number of years since they were born.
Correlation
- Two tailed hypothesis - There will be a relationship between how people perceive attractiveness as measured by a Bi-polar rating scale, and how old they are, as measured by the number of years, since they were born.
- Null hypotheses - There will be no relationship between how people perceive attractiveness, as measured by a Bi-polar rating scale, and how old they are, as measured by the number of years since they were born.
Strength of Correlation:
- allows researchers to investigate topics that would not be available otherwise due to it being unethical, such as looking at factors linked to child abuse.
Weakness of Correlation:
- it is difficult to produce a cause and effect as a correlation only looks at the relationship between variables and may not take all the other factors into account.
Related discussions on The Student Room
- MSc Forensic Psychology or MSc Criminal Investigation? »
- How can I get paid experience in Psychology/Childcare without a University Degree? »
- need ideas for my psychology final year project! »
- sister never supports me »
- Broca and Wernicke's work - Experiments or case studies? »
- Psychology Dissertation »
- Alternative careers/postgrad options with psych degree - Medical? »
- Can I do an Apprenticeship whilst studying a Level 3 BTEC? »
- 3rd Year Dissertation Ideas (ODP) »
- alternative careers with psych degree/postgraduate options- Medical? »
Comments
No comments have yet been made