This summary helped me get a 9 for my statistics exam. It is a summary of all the lecture notes and important information from the chapters of the Andy Field- Discovering Statistics book. It includes lots of pictures as well, to help you understand the concepts.
Summary Statistics
Background recap- Lecture 1
Types of data analysis
• Quantitative methods
- Testing theories using numbers
- Describing and analysing data using statistics
• Qualitative methods
- Testing theories using language: Interviews, conversations, newspapers and
media broadcasts
One is not better than another, they can complement each other because they give
different information.
Why use statistics?
• Impossible to know from every single person how they behave or appreciate certain
ads or products etc. → Do research on a sample taken from the population we are
interested in. These results do not necessarily apply to everyone.
• We use statistics so we can generalise our findings → How big is the chance that the
result you found for a small group (your sample) also applies to your entire
population?
• Researchers want to make big claims about big groups
• Representative groups are important
The research process: Setting up a research study
Statistics is only part of the research process.
Statistics will not help if your study is badly
designed. Set up your study properly:
1. You find something that needs explaining
or read a theory you want to test
(observation)
2. Based on your observation or theory you
have certain expectations
3. Formulate testable hypotheses
4. Test the hypotheses by collecting data
5. Evaluate the results: does it fit the
hypotheses
Results may give reason to revise the theory and
go through the steps once more.
,The research process: generating and testing theories
• Theory= A hypothesized general principle or set of principles that explain known
findings about a topic and from which new hypotheses can be generated.
• Hypothesis= A prediction from the literature, a theory or observation. Expected
relationship between variables.
The number of people turning up for a reality TV audition that have narcissistic
personality disorder will be higher than the general level (1%) in the population.
• Hypothesis needs to be falsified= Act of disproving a theory or hypothesis.
One falsification is stronger than an infinite number of confirmations.
Falsification
“All swans are white”
• You can never prove a hypothesis based on verifications because you don’t know
what you didn’t observe yet.
• It is possible to show a hypothesis can be rejected= falsification
Variables
To generate hypotheses and analyse data you need to define the variables you are
interested in.
• Variable= Anything that can be measured and can differ across entities or time. It
varies.
• Research describes and tries to explain variability
• During research you are interested in (the relationship between) theoretical
constructs. Operationalising theoretical constructs= Variables
- Type of ad: Ads with and without certain words
- Product colour: Products with one of three different colours
- Type of feedback: Language training with one of five types of feedback
Types of variables
• Independent variable/predictor variable= The variable you manipulate in
experiments
• Dependent variable= The variable that shows the result of the influence. Not
manipulated, but measured
Swear words in ads- Experiment
DV: Mean score on scale of ad appreciation
IV: Presence of swear words in ads (present absent)
• Control variable= Does not vary and is not manipulated
• Moderator variable= A variable that influences the effect of the independent
variable
Example:
A study using Dutch-accented and German-accented speakers of English to study the effect
of type of English accent on speaker appreciation.
IV: Type of accent (Dutch and German-accented)
DV: Speaker appreciation
Control variable: Age if only participants who were 20 years old took part
Moderator: Gender if the type of accent has an effect on speaker appreciation for women but
not for men.
,Example variables: Level of education, Group (with/without training), Age, Type of ad
(with/without swear words), Language of the ad (native/foreign language), Communication
style (direct/indirect).
Levels of measurement NOIR
Categorical (entities are divided into different categories):
• Binary variable= There are only 2 categories
Dead or alive
• Nominal variable= There are more than 2 categories
Whether someone is an omnivore, vegetarian, vegan or fruitarian
• Ordinal variable= Same as nominal variable but the categories can be ordered
Whether people got a fail, a pass or a distinction in their exam. One category is
better or higher than another.
Continuous (entities get a score):
• Interval variable= Equal intervals on the scale represent equal differences in the
property being measured
The difference between 6 and 8 is the same to the difference between 13 and 15
• Ratio variable= Same as interval variable, but the ratios of scores on the scale must
also make sense. Therefore, there must be an absolute zero point.
Someone earning 16 euros, earns twice as much as someone earning 8 euros.
Someone earning 0 euros, earns nothing.
Type of calculations you can do in statistics depend on the level of measurement. The
further you go in NOIR, the more measurements you can do with it in statistics.
Measurement error
• Measurement error= The discrepancy between the
actual value we’re trying to measure and the number
we use to represent that value.
You really weigh 80 kg, but when you stand on the scale
it says 83 kg.
We can keep measurement error low by using valid and
reliable measures:
• Validity= If an instrument measures what it was set
out to measure.
• Reliability= The ability of the measure to get the same
results under the same conditions at a later time.
No validity without reliability
Research design
• Correlational research
Observing what naturally goes on in the world without directly interfering with it
Surveys on opinions
• Cross-sectional research
When data come from people at different age points with different people
representing each age point.
• Experimental research
, - One or more variables is systematically manipulated to see their effect (alone or
in combination) on a dependent variable
- Different levels of control and randomization (randomly choosing participants)
- Statements can be made about cause and effect
Type of research depends on the research question.
Correlation is not causation!
Experimental research methods: Causation
Cause and Effect. 3 requirements:
• Cause and effect must occur close together in time
• The cause must occur before the effect
• The effect should never occur without the presence of the cause
• There should be no confounding variables= A variable other than the predictor
variables that potentially affects an outcome variable
Relationship between breast implants and suicide is confounded by self-esteem.
• You want to rule out confounds: An effect should be present when the cause is
present and when the cause is absent the effect should be absent as well.
Experimental research design: Methods of data collection
• Between-subjects design= Different entities in different experimental conditions →
More unsystematic variation, because people differ
• Within-subjects design= Same entities take part in all experimental conditions
Cheaper, but practice may help in improving and participants may become tired
To increase probability of finding an effect that does exist in the population, you
should increase sample size and measure the effect within subjects.
Whether you are using a between-subjects or within-subjects design affects your variation:
• Systematic variation= Differences in data caused by your manipulation.
• Unsystematic variation= Differences in data created by unknown factors (Gender,
age, IQ, time of the day).
Randomization of order of conditions and random assignment of participants to conditions
minimizes unsystematic variation.
When you run statistical analyses, you determine how much of the total variation in your
data is systematic and how much is not → Test statistic:
A control condition in the research design provides you with a reference point to
determine what change (if there was any) occurred when a variable was affected.
What statistics can and cannot do
What it can do:
• Data description
• Inferencing from sample to population
What does it mean if the hypothesis is not supported by the data?
• The effect or difference you are looking for does not exist
• The research design is not sensitive enough to pick up the difference that does exist
Statistics does not answer why→For the why you need qualitative research
Statistics does not provide information on causality
Voordelen van het kopen van samenvattingen bij Stuvia op een rij:
Verzekerd van kwaliteit door reviews
Stuvia-klanten hebben meer dan 700.000 samenvattingen beoordeeld. Zo weet je zeker dat je de beste documenten koopt!
Snel en makkelijk kopen
Je betaalt supersnel en eenmalig met iDeal, creditcard of Stuvia-tegoed voor de samenvatting. Zonder lidmaatschap.
Focus op de essentie
Samenvattingen worden geschreven voor en door anderen. Daarom zijn de samenvattingen altijd betrouwbaar en actueel. Zo kom je snel tot de kern!
Veelgestelde vragen
Wat krijg ik als ik dit document koop?
Je krijgt een PDF, die direct beschikbaar is na je aankoop. Het gekochte document is altijd, overal en oneindig toegankelijk via je profiel.
Tevredenheidsgarantie: hoe werkt dat?
Onze tevredenheidsgarantie zorgt ervoor dat je altijd een studiedocument vindt dat goed bij je past. Je vult een formulier in en onze klantenservice regelt de rest.
Van wie koop ik deze samenvatting?
Stuvia is een marktplaats, je koop dit document dus niet van ons, maar van verkoper lauralisa. Stuvia faciliteert de betaling aan de verkoper.
Zit ik meteen vast aan een abonnement?
Nee, je koopt alleen deze samenvatting voor €6,19. Je zit daarna nergens aan vast.
