Methodological solutions and statistical advice

PJSchmid Scientific Consulting can assist you in designing and improving your environmental or ecological studies in applied and scientific research. We also offer software solutions for the fractal and multifractal analysis of time series, profiles and surfaces (©EcoStatistics, release 2019). Our team can help you to improve the study design, sampling strategy and sample size to ensure the highest possible statistical precision. We can be consulted in the following areas of data collection and analyses1-6:

Study design

  • Testable hypotheses: scrutinizing ecological research-questions as statistically testable hypotheses.
  • Experimental design: assistance in designing testable field or laboratory experiments.
  • Sampling design: advice on quantitative and qualitative sampling methods7 (e.g., EU directives).
  • Statistical interpretation: assistance in the interpretation of statistical results.
  • Study outline: advice on making your data publishable in peer-reviewed journals.
  • Data presentation: assisting in the design of statistical data presentations for best public impact.

Sample size

An important statistical consideration for any kind of research is the justification of sample size and with it the data precision. Sample size is critical for correctly representing a population within a predefined margin of error. Determining sample size is a central issue because samples that are too small often lead to inaccurate results and wrong conclusions. Generally, as the number of samples increases, parameter estimates become more precise. We can advise you on the appropriate sample size for the planned statistical analysis.

Power analysis

The power of a statistical test can be defined as the probability that will lead to the rejection of the null hypothesis. We can advise you how to conduct a power analysis for your study, to estimate the sample size that you require or to validate that the sample size you have planned is sufficient. To do this, we either refer to any data you collected in a pilot study, or to relevant, published data.

Statistical inference

The most common purpose of statistical inference is hypothesis testing. To decide how likely the sample results match a hypothesis about a population, mathematical principles are used in the statistical analysis of data. You can consult us on the appropriate use of the following analysis and modelling techniques:

  • Descriptive statistics: standard and robust measures of location, spread, shape, entropy and dispersion.
  • Simulations: permutation, bootstrap, jackknife methods.
  • Hypothesis testing: nonparametric, parametric and exact test statistics.
  • Statistical modelling: standard and robust regression models, including model II regression.
  • Multivariate statistics: such as CA, DCCA, MDS, NMDS.
  • Graphical presentation: graphical data visualisation for publication and presentation.

Selected scientific publications

1. Schmid, P.E. & Schmid-Araya, J.M. 1997.  Freshwat. Biol. 38, 67  ( see abstract ).
2. Schmid, P.E., Tokeshi, M. & Schmid-Araya, J.M. 2000.  Science 289, 1557  ( see abstract ).
3. Schmid, P.E., Tokeshi, M. & Schmid-Araya, J.M. 2002. Proc. R. Soc. Lond. B 269, 2587  (download this publication).
4. Schmid-Araya, J.M. et al. 2002a.  J. Anim. Ecol. 71, 1056  ( see abstract ).
5. Schmid-Araya, J.M. et al. 2002b.  Ecology 83, 1271  ( see abstract )
6. Schmid-Araya, J.M. et al. 2016. Ecology 97, 3099  ( download this publication).
7. Schmid, P.E. & Schmid-Araya, J.M. 2010. Arch. Hydrobiol. 176, 365  ( see abstract).

For further information and questions related to statistical consultations, and software for fractal and multifractal analysis please contact us by email: