Lecture 1 - Experimental Design and Analysis Flashcards Preview

Statistics > Lecture 1 - Experimental Design and Analysis > Flashcards

Flashcards in Lecture 1 - Experimental Design and Analysis Deck (18)
Loading flashcards...
1
Q

What’s another name for the IV

A

the treatment variable

2
Q

What’s another name for the DV

A

the measure

3
Q

Which is experimentally manipulated, the IV or the DV

A

the IV

The DV changes because of the changes in the IV

4
Q

What are quantitative variables

A

they respresent variation in amount such as ‘amount of drug’ ‘loudness of stimuli’ etc

5
Q

What are qualitative variables

A

they represent variation in kind or type eg classical music, rock music, teaching strategy

6
Q

What are classification variables

A

they systematically vary characteristics which are intrinsic to the subjects of the experiment, such as age, gender, species, word type, IQ etc

7
Q

What is a nuisance variable?

A

Potential IV’s that if uncontrolled may have a SYSTEMATIC influence on the different treatment conditons
such as time of day, the researcher doing it etc

8
Q

What is a randomised block design?

A

uses blocks of subjects who are matched closely on a relevent attribute

Typically a subject is treated as a ‘block’ - they serve in all the treatment conditons of the IV

So this is basically a fancy way of talking about a type of repeated measures design.

9
Q

Describe what a statistical hypothesis is

A

Statistical hypothesis are a set of precise hypothesis about the parameters of the different treatment populations

10
Q

What is a (alpha)

A

the significance level (usually 0.05)

11
Q

What is a type 1 error?

A

We wrongly accept the alternate hypothesis and reject H0 = we see an effect where thee isn’t one

We may be willing to accept more type 1 errors when we want to discover new things

12
Q

What is a type 2 error?

A

We wrongly accepy H0 and reject the H1 = we fail to see an effect even though one is there

We may be willing to accept more type 2 errors when we don’t want to falsely assume something (e.g. safety of a medicine)

13
Q

What is a research hypothesis?

A

Research hypothesis are general statements about the assumed nature of the world that is then translated into creating an experiment

14
Q

What is H0?

A

the null hypothesis (its H with the 0 below it rather than on the same line, like a chemistry symbol)

15
Q

What does u1, u2, u3 mean (u with the long line on the left hand side)

A

they represent the results of the conditions (usually the mean)

16
Q

What is H1?

A

the alternate hypothesis

17
Q

What is an important limitation of the data collected in any experiment, with respect to the hypothesis?

A

The data collected tells us about a sample but the hypothesis aims to make a prediction about the whole population - this is why we must decide whether to accept or reject h0

18
Q

How do we decide whether to accept or reject H0?

A

We calculate a test-statistic based on the f-distribution value of our data. The value we adopt is called the significance level and is referred to as a (alpha). Usually our alpha is 0.05 in psychology.