Small Sample

Sample, Small

 

a statistical sample of such a small size n that it is impossible to apply to it the simple classical formulas that are valid only asymptotically as n —. co. The distinctive features of statistical estimates of parameters using small samples can be most easily understood using the example of a normal distribution, for which samples of size n ≤ 30 are usually considered small. Suppose it is necessary to estimate an unknown mean value a of a normal population with unknown variance σ2 using a sample x1, x2, … , xn. From the population we denote

In estimating a, we proceed from the fact that the probability distribution of the variable

is independent of a and σ.

The probability ω that the inequality —tω < t < tω holds and that the inequalities

equivalent to it hold is calculated here using the formula

where s(t, n — 1) is the probability density for the Student distribution with n — 1 degrees of freedom. By determining the corresponding tω for given n and ω (0 < ω < 1), which may be done, for example, by using tables, we obtain rule (1) for finding the confidence limits for a with a significance of level ω.

For large n, equation (2), which relates ω and tω, can be replaced by the approximate formula

This formula is sometimes incorrectly used for determining tω for small n, which leads to gross errors. Thus, for ω =0.99, using formula (3) we find that t0.99 = 2.58. True values of t0.99 for small n are given in Table 1.

Table 1
n2345102030
t0.9963.669.925.844.603.252.862.76

If formula (3) is used for n = 5, we may conclude that the inequality

is satisfied with probability 0.99. In fact, in the case of five observations, the probability of this inequality is equal to only 0.94, while the inequality

has probability 0.99, as can be seen from Table 1.

Similar methods of estimating the parameters of multivariate distributions (for example, correlation coefficients) using small samples have been developed.

REFERENCES

Cramér, H. Matematicheskie melody statistiki. Moscow, 1948. (Translated from English.)
Kolmogorov, A. N. “Opredelenie tsentra rasseivaniia i mery tochnosti po ogranichennomu chislu nabliudenii.” Izv. AN SSSR: Seriia matematicheskaia, 1942, vol. 6, nos. 1-2.
Bol’shev, L. N., and N. V. Smirnov. Tablitsy matematicheskoi statistiki. Moscow, 1965.

IU. V. PROKHOROV

References in periodicals archive ?
We agree that the small sample size is a definite limitation of our study, which not only limits drawing conclusions from it, but at the same time, overestimates the significance of p-value.
This textbook/manual will be useful to those conducting research in a real-life healthcare setting with limited funds and small sample sizes.
A small sample is drawn into a tube by vacuum suction.
The study evaluated the effectiveness of log-linear presmoothing (Holland & Thayer, 1987) on the accuracy of small sample chained equipercentile equatings under two conditions (i.
The second point of concern is the very small sample size used in this study.
The small sample size and rapid test time associated with this new blood glucose meter may make testing less painful and could encourage those with diabetes to test more often.
Given the small sample size, generalizing beyond the class may be difficult.
Sample preparation can require a lot of time and expertise, particularly when dealing with small sample amounts, according to the company.
McDougall and Smith's article gives practitioners advice on how to conduct studies with small sample sizes, and Sink and Stroh provide help with determining the practical significance of research studies through the use of effect-size estimates.
Qualitative researchers using small sample N's use a different approach in selecting samples than researchers concerned with extending generalizations to other populations (LeCompte & Preissle 1993).
The Library of Congress has already compiled a small sample.

Full browser ?