Unit 3 Flashcards
replacement
avoiding using animals
reduction
using fewer animals
refinement
change the way the experiment is carried out so animals suffer as little as possible
pilot study
small scale investigation of a planned research project used to help
- develop and/or practice protocols in order to ensure validity of experimental design
- check effectiveness of techniques
- find a suitable range of values for the independent variable
- identify and control confounding variables
- identify suitable numbers of replicates
types of variables
independant- variable that is changed or controlled
dependant- variable that is being tested or measured
confounding- any factor that affects dependent variable that isn’t independent- must be held constant or monitored
randomised block design
used in cases where confounding variables can’t easily be controlled, blocks or experimental and control groups can be distributed in such a way that the influence of any confounding variable is likely to be same across all groups
what types of data can discrete or continuous variables give rise to?
quantitative, qualitative or ranked data
limitations and use of simple (one independent variable) experimental design
findings may not be applicable to wider setting, control of lab conditions means that these experiments can be conducted more easily than in the field
multifactorial experimental design
more than one independant variable
in vivo
carried out using living organisms, tend to be more complex than in vitro experiments
in vitro
carried out using biological molecules outwith normal biological system ie. cell culture
observational studies
studies where there is no independent variable, good at detecting correlation but as they don’t directly test the model, not useful for determining causation
control groups are used for
comparison with treatment results
negative control
provides results in the absence of a treatment
positive control
treatment that is included to check that the system can detect a positive result when it occurs
when is a representative sample of the population selected?
where it is impractical to measure every individual
what determines appropriate sample size?
extent of natural variation within a population, more variable= larger sample size
a representative sample should share the
same mean and degree of variation about the mean as the population as a whole
random sampling
members of the population have an equal chance of being selected
systematic sampling
members of a population are selected at regular
intervals
stratified sampling
population is divided into categories that are then
sampled proportionally
reason for variation in experimental results
due to reliability of measurement methods and/or inherent variation in the specimens
how to determine the reliability of measuring instruments or procedures
by carrying out repeated measurements or readings of an individual data point, the variation observed indicates the precision of the measurement instrument or procedure but not necessarily its accuracy
how can natural variation in biological material used be determined?
by measuring a sample of individuals from the population- mean of these repeated measurements will give an indication of the true value being measured
repeating experiements checks
reliability of results
scientific cycle
construction of a testable hypothesis, experimental design, gathering, recording, analysis of data, evaluation of results, conclusions and the formation of new hypotheses where necessary
null hypothesis
hypotheses stating that one variable will have no effect on the other
why should publication of methods, data, analysis and conclusions in scientific reports be reported in standard format and in detail?
so that others are able to repeat the experiment
review articles
summarise current knowledge and recent findings in a particular field
what do scientific ideas need before being published?
independent verification and must be supported by similar findings in a number of other investigations
all findings are subject to critical evaluation from
peers in peer review and the public through the wider media
integrity and honesty in scientific literature
unbiased presentation of results, citing and
providing references, avoiding plagiarism
when using humans what should be ensured?
informed consent and subjects must be aware of the right to withdraw data and confidentiality
what can influence scientific research?
legislation, regulation, policy and funding
Scientific reports should contain
an explanatory title, a summary including aims and findings, an introduction explaining the purpose and context of study including the use of several sources, supporting statements, citations, and references, method section containing sufficient information to allow another investigator to repeat the work
Experimental design should test
the intended aim or hypothesis
what should treatment effects be compared to
controls and any confounding variables
what is necessary to ensure representative sampling and allow the investigator to show that altering the independent variable has, in fact, had an effect on the dependent variable?
a large and unbiased sample group
how is data usually presented?
as tables and graphs, with statistical analysis used as appropriate
statistical analysis
median, mode, mean
standard deviation
used to assess the spread of the data
what type of results are unlikely to be due to chance alone?
statistically significant results
what do confidence intervals and error bars suggest?
variability of data around a mean, if the treatment average differs from the control average sufficiently for their confidence intervals not to overlap then the data can be said to be different
conclusions should refer to
aim, results and hypothesis and existing knowledge and the results of other investigations
what should be taken into account to decide if conclusions are valid?
validity and reliability of experimental design, may include discussing controlled variables, sample
size, repetition and accuracy of measurements
what should consideration be given to pertaining results?
whether the results can be attributed to correlation or causation
