Data collection forms an integral part of biological statistics and a core step in the conduct of experiments. Collection techniques vary according to the aim and objectives of the experiment. The various techniques available to researchers include:
This is a technique where researchers directly gather data by measuring/counting/weighing variables using appropriate instruments. For example, collecting data on the weight of students in a particular school would require going to the school with a weighing balance, selecting samples among the students, taking and recording their weights.
In this method, researchers obtain data by communicating with the subjects of study. It involves socially interacting with humans. Data can be collected via face-to-face or telephone interviews, direct or mailed questionnaires, or any other means involving human social interactions.
A variable can be defined as any measurable, attributable, or rank-able factor that can vary during experiments. Examples of variables include colour, weight, height, number of leaves, class of degrees, etc. Choosing the appropriate statistical test for your data depends on the kind of variables measured in the course of the research.
The type of variable is often determined by the type of data measured during collection. If a variable consists of quantitative data, such a variable is also said to be quantitative. This also means that quantitative variables can be continuous of discontinuous just like data. Hence, volume, weight, and distance would be considered continuous variables while the number of leaves, number of flagella, and number of teeth would be considered discontinuous variables.
A variable that consists of data that cannot be quantified but categorized is known as a categorical variable. Categorized variables can be binary, nominal, or ordinal. Binary variables are those that are recorded in either of two ways, examples, yes/no, present/absent, etc. Nominal or attribute variables are categorized without any specific rank or order, examples, colour, brands, etc. Ordinal variables are those that are ranked or arranged in specific orders. Examples include the class of degrees, positions in a race, etc.
Variables can also be named based on the role they perform during experiments. Some variables are meant to cause effects while some are meant to measure the effects caused by the former. The ‘cause’ variables are referred to as the independent variables, whereas, those that measure effects are referred to as the dependent variables.
Some other variables, however, are held constant throughout the course of experiments so as to be able to isolate the cause and effects’ variables. Such variables are referred to as constant or controlled variables. As an example, let us look at the illustration below:
A student wanted to measure the effects of temperature on the growth rate of a plant species. He acquired 30 seedlings of the plant and divided them into 3 groups. He then planted the seedlings in similar pots containing the same soil. He allowed the first group to grow at room temperature, the second group at 30 degrees, and the third group at 35 degrees. He measured the height of each plant at two days intervals. What is the dependent, independent, and the controlled variable?
In this case, the variable that produces the effect would be the temperature. The effect would be measurable on the growth-indicating factor of the plant – the height. Hence, the independent variable would be the temperature while the dependent variable would be the height of the plants. All other conditions of growth like the size of the pot, the quantity of soil in each pot, the amount of light the plants are exposed to, etc., must be kept constant and they together represent the controlled or constant variable.
Avoiding or minimizing errors in the course of gathering data is very important in order to ensure that correct decisions are arrived at as far as the aim and objectives of the research are concerned. The correctness of a conclusion is dependent on the accuracy and the precision of the data collected.
Accuracy is the closeness of measurements to their true values while precision is the closeness of repeated measurements. That repeated measurements are close does not necessarily mean that the measurements are accurate. For example, a student that consistently measures the volume of water from a measuring cylinder by taking readings from the upper meniscus would have precise data but would be far from being accurate. There are different types of errors that can make measurements to be inaccurate:
