Inferential Statistics

Probability: The probability of the results of a study happening by chance and a one off. Its measured using the P Value. The smaller the P value the more reliable the results. A P value of P<0.05 is the normal value psychologists use, this means there is 5% chance of it being chance.

Significant: If your test is significant then you can reject your null hypothesis.

Levels of Measurement:

• Nominal: The data is put into catergories for example sorting people by height, tall, medium, short.
• Ordinal: Data is ordered in some way e.g alphabetically
• Interval: Data is measured in units of equal interval such as counting correct answers.

Type 1 Error: Rejecting the null hypothesis when you shouldn't,P value is too high

Type 2 Error: Accepting the null hypothesis when you shouldn't, P value is too low

The Hypothesis predicts a correlation between two sets of data.

Experimental Design: Matched Pairs

Levels of measurement: Ordinal/Interval

Step1: State the hypothesis and null hypothesis

Step2: Record the data and work out the difference

Step3: Find the observed value of Rho

Step4: Find the critical value of Rho

Step5: State the conclusion

The Hypothesis predicts a difference between two correlations

Experimental Design: Independent Measures

Levels of measurement: Nominal

Use a contingency table to plot data

Calculate the DF. DF = (Rows - 1) X (Columns - 1)

Work out the critical value.

If the Observed Value is less than the Critical value then accept the null hypothesis.

The Hypothesis predicts a difference between two sets of data

Experimental Design: Independent Measures

Levels of measurement: Ordinal/Interval

Step1: State the hypothesis and null hypothesis

Step2: Record the data and allocate points

Step3: Find the Observed value

Step4: Find the Critical Value

Step5: State the conclusion

The Hypothesis predicts a dfference between two sets of data

Experimental Design: Matched Pairs/Repeated Measures

Level of measurement: Ordinal/Interval

Step1: State the hypothesis and null hypothesis

Step2: Record the data and calculate the difference between scores and rank

Step3: Find the Observed value

Step4: Find the Critical Value

Step5: State the conclusion

Measures of central tendency: These try and calculate the average, it used on of these three methods

The Mean: Add up all of the scores and divide by the number of scores. Its not good for nominal data.

The Median: Its the middle value in an ordered list. Its not appropriate for nominal data.

The Mode: Its the value which is most common in the data. It is appropriate for nominal data, or data in categories.

Measures of dispersion: Trys to tell us about the spread of the data

The Range: Calculated by finding the difference between the highest and lowest score but might be effected by extreme values.

Standard Deviation: Expresses the spread of data around the mean, all bits of data are examined.

Spearman's Rho: A test of correlation is needed as the hypothesis predicted a correlation and the data collected is ordinal

Chi-Square: As the data has been put into categories and is nominal data, the experiment design was also independent groups.

Mann-Whitney: A test of difference is needed between two groups. Uses independent groups and uses ordinal data.

Wilcoxon: A difference is predicted between two conditions. It will use repeated measures and interval data.