Evaluation of practicals Flashcards
What is Reliability
o Measure of the closeness of agreement between individual results
o It is a measure of the ‘repeatability’ of the data collection pricess
o However as shown previously, closeness of replicates does not mean that the data are accutate
- Repeatable results
o Are replicate results that are in close agreement
o You can use mathematical methods to describe variation in relaicte reulsts
Reproducible
- results that can be reproduced by someone else following exactly the same procedure
- You only comment on this in the evulative task if you are given results from other students.
- Precision
o Refers to the number of significant fugres or decmal places in readings
- Accuracy
o Meaure of the closeness of agreement between individual results or a set of results and an accepted true value
o In biology it is often difficult to know the ‘true ‘ value’
- Anomalous reulsts
o Results that do not fit the trend
o They are sometimes know as outlers
o They can be
♣ Replicatre results that differ significationly from others
♣ A results (which may ot may not be a mean ) for one value of the independent variable that does not fit the overall trend and is not included in the curve of best fit.
- Validity
o Confidence that you can have in your conclusions
o In a valid investifation
♣ You have measured what you set out to measure
♣ Changing the independent variable led to changes in the dependet variable which you have mreaure.d
What should you consider when assessing validity
- If asked to comment on the validity of an investigation then you should consider the following
o The limitations in the procedure
o Any uncontrolled variables
o The effects of errors (systematic and random ) on the results
o The reliability and repeatability of the results
o The precision of the data collected
o The accuracy of the reults
What two errors may have occurred in the procedure? What are they
- Systematic erors
o Same throughout the investigation
o A common systematic eror is that the measuring device may give readings that are out by a certain value. It could be that one of the controlled cariables is always incorrect by the same quantity. If there are small systematic erors then the data may not be precies, but not accurate. The effect is to oversestimate or underestimate the true values
o Random errors
♣ Occur when you do not carry out the procedure exactly the same way each time. You may also read the apparatus in a slightly different way.
• These errors affect some of the results, butnot all of them .they do not always affect the result in the same way. Random errors could also be the result of variation in biological material.
Name 12 common limitations
One measurement or only two replicates
Difficulty in judging colours or looking for colour changes when using by eye methods
Insufficient intermediate readings taken . every 60 seconds for example
Insufficient number of values of independent variable e.g. only at 20oc , 30oc , 40oc, 80oc 100oc
Insufficient range of independent variable 30oc to 50oc
Inconsistent stirring of solutions e.g. reaction mixture not stirred same way before taking smaples
Samples taken from different sources e.g. potato cylinders from different potatoes
Method of timing e.g. stop clock
Impossible to start all the reactions at the same time
pH not controlled e.g. in an enzyme experiemtn
How does using only one or two measurements affect the results? How may this be fixed?
Unable to detect anomalous reults thus decreasing reliability
Perform at least three replicates
How does “Difficulty in judging colours or looking for colour changes when using by eye methods” affect the results? How may this be fixed?
Lack of consistency in judgement may lead to less precise , less reliable and less accurate results
Use color standards for comparison, use a colorimeter
How does “Insufficient intermediate readings taken . every 60 seconds for example” affect the results? How may this be fixed?
Lack of results between the intervals – could miss changes so making conclusions about trends less valid
Take readings/ measuremetns at smaller intervals e.g. at 30 not at 60s
How does “Insufficient number of values of independent variable e.g. only at 20oc , 30oc , 40oc, 80oc 100oc” affect the results? How may this be fixed?
Lack of results between the intervals – could miss changes so making conclusions about trends less valid
Include more intermediate temperatures within the range e.g. 25oc, 35 oc, 45oc , 50oc etc
How does “Insufficient range of independent variable 30oc to 50oc” affect the results? How may this be fixed?
Lack of resuts beyond the range –investigated – could make it difficult to identify trend or pattern
Extend the range e.g. 10oc to 80oc
