Statistical test

Página 1

Descriptive - measures of central tendency, measure of disposition

Significance - how likely it is that that we are wrong.

We try to generalise from sample of population

sample errors happen a lot

p_> 0.5 (95% of being right) the level of significance is set to 0.05 or 5% this is expressed or 5% P is the likelihood of getting the findings if they null hypothesis is true. Anything less that 0.05 is described as significant it is unlikely that the null hypothesis is true. Anything above 0.05 is not significant as we can reject the null hypothesis. Null hypothesis - no link or correlation between 2 co-variables. If research is significant this will be rejected, if not, it'll be the experimental hypothesis will be accepted.

Type 1 error - if we reject the null hypothesis and say there is an offer and we are wrong and we have made a type 1 error. The probability of making a type 1 error is the level of significance in the stats test.

Type 2 error - if we reject the they-thesis and we say there is no effect and we are wrong then we have made a type 2 error. Probibiltiy of making a type 2 error is unknown but can be caused by a small sample size.

question - what is it and what's the likely hood of making a type 1 error in the stem } level of significance.

Página 2

Levels of measurement - ordinal } rank in order EG: initial - scores (how many words remembered

Sign test

2 categories

positive and negative (did or didn't)

Ordinal } doesn't have to have equal intervals.

Rank in order } agree disagree, number or words remembered,

What grade did you get in GCSE's

Interval data } measured exactly - equal intervals

Time, height, weight

We can measure and compare the exact time because the intervals on a ruler are equal.

Página 3

Tests difference

Chi square

Nominal data (categories)

Independent measures

frequency table } a number is made of degrees of freedom (df)

Calculation is df = (R-1)x(C-1) HAVE TO WORK OUT

Wilcoxon signed rank

ordinal data

repeated measures

Man whitney U

ordinal data

independant measures

Sign test

nominal

repeated measures

Indépendant t

interval or ratio data

indépendant measure

Related t

interval

on a scale but has no true zero

weight, time and distance

Página 4

To what extent do you believe that you get stressed 1-10? What do you feel you get stressed about at work?

Do you smoke Yes/No? How do you feel towards smoking?

Do you have sleep problems? How many hours sleep at night?

Are you angry? How angry do you feel?

Interval

Nominal

Ordinal

Interval

Ordinal

Interval

Nominal

Nominal

Nominal

Interval