Analytical 2, chemometrics

Question 1

what is chemometrics

Answer 1

the chemical discipline that uses mathmatical and statistical methods to design or select optimal measurements and experiments. basically data production and collection

Question 2

statistical population

Answer 2

set of all cases or measurements that are the objects of the statistical study

Question 3

statistical sample

Answer 3

a sample is a statistically significant portion of the population

Question 4

accuracy

Answer 4

consistency between the result of a detrmination xi or the mean of n results (xbar) provided by a given method and the value hels as true (x’)

Question 5

precision

Answer 5

consistency between results obtained by independently subjecting identical aliquots of the same sample to the same analytical process to determine the same analyte

Question 6

precise but not accurate

Answer 6

target—- all darts together but not near the middle

Question 7

accurate but not precise

Answer 7

target —-darts not together but all near the middle

Question 8

not accurate and not precise

Answer 8

target—- darts scattered around target, not together and at various distances from the middle

Question 9

accurate and precise

Answer 9

target—- all darts together and on the middle of the target

Question 10

hypothesis testing - one tailed test - when do we do it

Answer 10

to test whether a sample mean is either greater than or smaller than a certain value, but not both
eg is method a more accurate than method b

Question 11

hypothesis testing- two tailed test - when do we do it

Answer 11

to test whether there is a significant difference bewteen the mean sample and the known value, regardless of sign or difference.
eg is there a significant difference between the mean values, are both methods accurate
what is the purity of the sample

Question 12

hypothesis testing : steps

Answer 12

1: null hypothesis is defined. postulates that two or more observed quantities are the same
2: statistic test must be chosen ie one or two tailed test
3: significance level must be defined
4: decision rule must be set up that is based on signifcance level and distribution of test statistics
5: calulation of test statistic and decision on the decision rule must be made

Question 13

if we havent been told what percentage to test to in a null hypothesis we presume a significance level of…

Answer 13

5% aka 0.05

Question 14

not rejecting h0 (t tab > t exp) implies

Answer 14

there is no experimental evidence to assert that xbar is significantly different from u for a given alpha value (method is accurate)

Question 15

rejecting h0 (t tab < t exp) implies

Answer 15

that xbar is significantly different from u for a given alpha value (method is not accurate)

Question 16

calculation of t exp

Answer 16

t exp= l xbar- u l rootn/s

Question 17

decision rules:
if t exp> t alpha, v

Answer 17

h 0 is rejected (method not accurate, there is systematic error)

Question 18

decision rules:
if t exp< t alpha, v

Answer 18

h 0 is retained (no evidence of systematic error. accurate method)

Question 19

v in hypothesis testing equations means

Answer 19

number of degrees of freedom (n-1)