For environmental assessments, and in particular, biological assessments, data are measurements and statistics taken from a population or source. Generally, the purpose of data collection in the field of biological assessments is to provide information on biological attributes and answers to ecological questions.
The FVC team collects data for most of the environmental consulting projects we carry out. Such projects typically involve personnel mobilising to the field to record data relating to biological features within a given area. Simply put, what plants, vegetation (collections of plants), animals and interactions between these (ecosystem) exist during that snapshot in time?
Once interpreted, this data can help us to inform our clients of the baseline biological values in their project area, or how the values may have changed since the last time data was collected from that same location, in the case of monitoring programs. Either way, robust data that is collected for specified parameters, in a defined way, then analysed and interpreted by experienced environmental consultants, provides important information about the living systems we are tasked with assessing.
Data defined
Data are the building blocks of knowledge, providing a snapshot in time of a population or source. Data itself are just a collection of measurements, but when transformed, analysed or interpreted they provide an image of the subject and can help explain the natural world.
Data can be qualitative or quantitative. Qualitative data are more descriptive and observational, such as written observations or ranks, while quantitative data are numerical and measurements of attributes, such as counts. Both have their uses, benefits and short-comings. Qualitative data can make sense of the natural world that cannot be easily quantified, while quantitative data can be calculated, analysed, and interpreted.
The purpose of data
Data are used for many purposes, including providing baseline information, interpreting changes over time, and informing management activities.
Fundamental to understanding a biological source is robust, quality data collection. Interpretation of data analysis, and subsequent outcomes or decisions, are based on the underlying data.
Data user beware!
Data collection and subsequent conclusions can be influenced by problems such as bias, under-representation and inadequacy.
One of the most important aspects of data collection is use is having well-defined methodologies for data sampling, recording and analysis. Repeatable data collection is fundamental to any scientific study. This can include collecting data from spatially defined plots such as pegged quadrats in the case of monitoring programs. Typically, collecting robust data will also involve the use of data collection tools such as electronic devices and software or apps that improve efficiencies in the field and provide good, clean data ready for analysis. Importantly also, repeatable methodologies will incorporate the use of statistical analysis software, which finally will require interpretation by suitably qualified and experienced practitioners, to compile meaningful results and conclusions.
The pitfalls of data bias
Data collection must be free from bias (random), systematic, representative, and adequate. Bias is anything that skews the conclusion from the fact. Bias may be unavoidable, accidental, or deliberate. For example, when collecting data on rehabilitated vegetation to assess success towards completion.
Data collection should be random, assuming that an appropriate number of samples will provide a snapshot of the state of the rehabilitation. If a practitioner were to only select areas of either good or bad rehabilitation, it would skew the result. Or, if comparing the species assemblages within the same area over multiple periods, it could bias results by sampling in different seasons, as some flora and fauna will have specific periods of presence and absence. There are methods to alleviate biases, and it is important to be aware of potential bias and design appropriate studies.
The importance of adequate sampling
To ensure a biological source is completely understood, data collection needs to be representative and adequate.
There is variation in any population, therefore, when collecting data, it is important to sample enough of the population to capture that variation. When collecting data, an adequate number of samples distributed over the population should provide robust data. How many is adequate? This can be difficult to determine, but there are statistical methods that provide an estimate of the number of samples required to facilitate robust conclusions.
It is important to be aware of the implications of too little data, as conclusions based on these may not be accurate.
The use of data to compare systems over time
Systematic data collection allows for comparisons over time and between comparable sources. If a practitioner wishes to compare the results of their study with others, using the same method of data collection provides more robust results than two different methods.
For example, vegetation of the Swan Coastal Plain was previously studied and categorised with an extensive survey published in 1994. To compare current surveys with this past data, the method of data collection must be consistent with the surveys that contributed to the 1994 publication. However, bias can creep into comparisons because many current surveys are not as extensive (less survey effort, therefore less representative) as that previously undertaken, so results must be interpreted with caution.
The power of interpreted data
Data itself is not the answer. Analysis and interpreted results tell the story.
Data analysis can be simple or complex, depending on the question asked and type of data available. Numerous statistical programs can transform and manipulate data to provide mathematical results. However, it is important to know and meet the underlying assumptions required. It is also important to understand the constraints of data. Of utmost importance is carrying out data analysis in a meaningful way and the methods required to be followed for interpreting results.
Anyone can input data into a program and press ‘calculate’, but it may not provide an accurate result.
The dish on data
Fundamental to the suitable interpretation of data is the underlying robustness of collection. Many things can influence the quality of data, including bias, thoroughness, and adequacy. Data analysis can be complex or simple but requires understanding the underlying assumptions and requirements. And whilst anyone can input data into a program and get a result, the accuracy of that result depends on the data behind the analysis and reliance the assumptions have been met during analysis.