Statistics

Question 1

What is standard deviation

Answer 1

It is the measure of spread of data around the mean. It’s the distance from mean to the point of inflection (concave to convex)
Within 1sd - 68% data
Within 2sd - 95% data

Question 2

Advantages of using standard deviation over range

Answer 2

SD gives you the spread of data around mean while range is just difference between highest and lowest value
SD reduces effect of anomalies while range includes anomalies
SD can be used to indicate if difference between results are significant

Question 3

How can knowing standard deviation help come to conclusion? In terms of overlap? And non

Answer 3

When you +/- sd from mean and there is overlap - it indicates that it’s likely that there is no significant difference between means
When you +/- sd from mean and there is no overlap - it indicates it’s likely that there is significant difference between means

Question 4

When is students t test used? How to write conclusion

Answer 4

When looking for difference between 2 means
-If calculated value is greater than or equal to critical value - Reject null hypothesis
There is less that 5%/0.05 probability that results are due to chance
-If calculated value is less than the critical value - Accept null hypothesis
There is greater that 5%/0.05 probability that results or difference are due to chance

Question 5

When is correlation coefficient statistic used? Conclusion? Null hypothesis?

Answer 5

-if investigating correlation or association between 2 continuous variables eg- effect of temperature of enzyme activity
NULL HYPOTHESIS- There is no significant correlation between two continuous variables(depending on context of question)
-If calculated value is greater than or equal to critical value - Reject null hypothesis
There is less that 5%/0.05 probability that correlation is due to chance
-If calculated value is less than the critical value - accept null hypothesis
There is greater that 5%/0.05 probability that correlation is due to chance

Question 6

Conclusion by analysis of p value

Answer 6

P value - probability that the results( context - correlation, difference which could be increase or decrease) are due to chance
And the value which is p= eg 0.05 means 5% probability or 95% confident that the results are significant
If p value is less than 0.05 - less than 5% probability that results are due to chance - there is significant difference - reject null
If p value is greater than or equal to 0.05 - greater than 5% probability that results are due to chance - there is no significant difference - accept null

Question 7

Which stats stats to use when? Justify

Answer 7

If measuring continuous variable -
1- investigating association/correlation between 2 variables - correlation coefficient
2- investigating differences in mean - students t test
If measuring frequencies(no. Of individuals in a category) and their difference - chi squared test