BIO3011 Lecture Notes - Lecture 1: Bar Chart, Scatter Plot, Normal Distribution
1. Variables and data
• Variable: characteristic measured on individuals drawn from a population under study
Types of variables:
o Response = dependent (variable we are interested in examining – responds to
manipulation) I.e. usually numerical
o Explanatory = independent (variable we are manipulating or measuring in order to
observe its effect on the response variable)
• Data: measurements of one or more variables made on a collection of individuals
• Population = total number of individuals that are used to summarise/describe a group of
measurements eg. mean, median, sd, se
o Parameters = summary describing the population eg. true mean
o Population parameters are constant
• Sample = much smaller set of individuals from the population (is representative of the
population)
o Statistics/estimate = approximation (estimate) of the truth – is subject to error
-estimates value of the true body size
-uses statistics to determine how good our estimates are
o The larger the sample size = the more certainty
o Estimates are random variables
o Properties of a good sample:
1. Independent selection of individuals eg. number each individual then choose
random numbers
2. Random selection of individuals
3. Sufficiently large
• Random sampling = each member of a population has an equal and independent chance of
being selected
• Bias = systematic discrepancy between estimates and the true population characteristic
o Volunteer bias = volunteers for a study are likely to be different, on average, from the
population
eg. volunteers for medical studies may be sicker than the general population
• Sampling error = difference between the estimate and the average value of the estimate
i.e. the difference between an estimate and the population parameter are being estimated by
chance
• Larger samples on average will have smaller sampling error
find more resources at oneclass.com
find more resources at oneclass.com
Document Summary
Variables and data, variable: characteristic measured on individuals drawn from a population under study. Uses statistics to determine how good our estimates are: the larger the sample size = the more certainty, estimates are random variables, properties of a good sample: Larger samples on average will have smaller sampling error: two most common descriptions of data, location (central tendency) Tell us about the average or typical individual. Median = middle measurement in set of ordered data. Mode = most frequent measurement: spread (variation) Tells us how variable the measurements are from individual to individual (how different the individuals are) Gives us perspective: how large are the differences between groups compared to variation with groups. Range = max min (biased small samples tend to give lower estimates of the range than large samples) Standard deviation = positive root of the variance. Samples are not independent: type i error = rejecting a true null hypothesis.