The phrase "statistical representativeness" is grammatically correct and can be used in written English.
This phrase is usually used to describe how closely a sample of data represents the population from which it was taken. For example, in order to make sure their survey was statistically representative of the population, the research team used a random sampling method.
For methodological reasons (sample size and lack of statistical representativeness), it is difficult to take these dimensions into account in the qualitative survey.
Statistical representativeness is then becoming less problematic in this type of field research (see for a justification and details, Yin 1984).
Given the enforced lack of true random sampling, we cannot be sure of the statistical representativeness of our groups of interviewees.
Considering the general objective of the analysis and the sampling methodology of Italian FADN database, which ensures a statistical representativeness at regional level (Abitabile and Scardera 2008), the correspondence between the sample and the universe is only evaluated in a qualitative way.
Whereas the goal of survey research based on random samples is statistical representativeness, the goal of multiple-case studies is quite different: "In a case model…the collection of units is, by design, not representative; each unit has its own probability of selection; and different units are subject to different questionnaires….
Here's my two cents: all this debate about statistical representativeness, small datasets, only a handful of companies, and sites being chosen by the gas companies misses a key result -- perhaps the most important one.
The statistical representative volume element is introduced to determine the minimum volume which provides the representativeness.
The variety, volume and velocity of the data may have changed, thus making the discipline more data driven than previously, but the fundamental statistical issues about representativeness and uncertainty in trade-off with practicability, ethics and privacy remain the same.
Therefore, samples from the two cities were merged to increase the statistical power and representativeness of study results.
The sample size was determined with the objective to ensure high statistical reliability and representativeness; the 95% CI should not exceed 5% of a key indicator estimate.
Cohort studies are vulnerable to the effects of attrition, reducing the available population from which inferences can be made and compromising the capacity to fit models with adequate statistical power and representativeness.
