Designation: E105 − 16 An American National Standard

Standard Practice for

Probability Sampling of Materials1
This standard is issued under the fixed designation E105; the number immediately following the designation indicates the year of
original adoption or, in the case of revision, the year of last revision. A number in parentheses indicates the year of last reapproval. A
superscript epsilon (´) indicates an editorial change since the last revision or reapproval.

1. Scope 4. Significance and Use

1.1 This practice is primarily a statement of principles for 4.1 The purpose of the sample may be to estimate properties
the guidance of ASTM technical committees and others in the of a larger population, such as a lot, pile or shipment, the
preparation of a sampling plan for a specific material. percentage of some constituent, the fraction of the items that
fail to meet (or meet) a specified requirement, the average
2. Referenced Documents characteristic or quality of an item, the total weight of the
shipment, or the probable maximum or minimum content of,
2.1 ASTM Standards: say, some chemical.
E122 Practice for Calculating Sample Size to Estimate, With
Specified Precision, the Average for a Characteristic of a 4.2 The purpose may be the rational disposition of a lot or
Lot or Process shipment without the intermediate step of the formation of an
E300 Practice for Sampling Industrial Chemicals estimate.
E141 Practice for Acceptance of Evidence Based on the 4.3 The purpose may be to provide aid toward rational
Results of Probability Sampling action concerning the production process that generated the lot,
E456 Terminology Relating to Quality and Statistics pile or shipment.
E1402 Guide for Sampling Design
4.4 Whatever the purpose of the sample, adhering to the
principles of probability sampling will allow the uncertainties,
3. Terminology
such as bias and variance of estimates or the risks of the
3.1 Definitions: rational disposition or action, to be calculated objectively and
3.1.1 For general terminology, refer to Terminology E456 validly from the theory of combinatorial probabilities. This
and Guide E1402. assumes, of course, that the sampling operations themselves
3.1.2 judgment sampling, n—a procedure whereby enu- were carried out properly, as well. For example, that any
merators select a few items of the population, based on visual, random numbers required were generated properly, the units to
positional, or other cues that are believed to be related to the be sampled from were correctly identified, located, and drawn,
variable of interest, so that the selected items appear to match and the measurements were made with measurement error at a
the population. level not exceeding the required purposes.
3.1.3 probability sampling plan, n—a sampling plan which 4.5 Determination of bias and variance and of risks can be
makes use of the theory of probability to combine a suitable calculated when the selection was only partially determined by
procedure for selecting sample items with an appropriate random numbers and a frame, but they then require supposi-
procedure for summarizing the test results so that inferences tions and assumptions which may be more or less mistaken or
may be drawn and risks calculated from the test results by the require additional data which may introduce experimental
theory of probability. error.
3.1.3.1 Discussion—For any given set of conditions, there
will usually be several possible plans, all valid, but differing in 5. Characteristics of a Probability Sampling Plan
speed, simplicity, and cost. Further discussion is provided in 5.1 A probability sampling plan will possess certain char-
Practice E141. acteristics of importance, as follows:
5.1.1 It will possess an objective procedure for the selection
of the sample, with the use of random numbers.
1
This practice is under the jurisdiction of ASTM Committee E11 on Quality and 5.1.2 It will include a definite formula for the estimate, if
Statistics and is the direct responsibility of Subcommittee E11.10 on Sampling / there is to be an estimate; also for the standard error of any
Statistics.
estimate. If the sample is used for decision without the
Current edition approved April 1, 2016. Published April 2016. Originally
approved in 1954. Last previous edition approved in 2010 as E105 – 10. DOI: intermediate step of an estimate, the decision process will
10.1520/E0105-16. follow definite rules. In acceptance sampling, for example,

Copyright © ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959. United States

1
 E105 − 16
these are often based on predetermined risks of taking the tion to size, will satisfy the requirement; likewise every plan
undesired action when the true levels of the characteristic will do so where the sample is made up of separate identifiable
concerned have predetermined values; for example, acceptable subsamples that were selected independently and by the use of
and rejectable quality levels may be specified. random numbers.
5.2 The minimum requirements that must be met in order to 6.5 A probability sampling plan for any particular material
obtain the characteristics mentioned in 5.1 appear in Section 6, must be workable, and if several alternative plans are possible,
which also indicates the minimum requirements for the de- each of which will provide the desired level of precision, the
scription of a satisfactory sampling plan. plan adopted should be the one that involves the lowest cost.
6.6 A probability sampling plan must describe the sampling
6. Minimum Standards for a Probability Sampling Plan units and how they are to be selected (with or without
6.1 For a sampling plan to have the requirements mentioned stratification, equal probabilities, etc.). The sampling plan must
in Section 5, it is necessary: also describe:
6.1.1 That every part of the pile, lot, or shipment have a 6.6.1 The formula for calculating an estimate (average
nonzero chance of selection, concentration, minimum concentration, range, total weight,
6.1.2 That these probabilities of selection be known, at least etc.),
for the parts actually selected, and 6.6.2 A formula or procedure by which to calculate the
6.1.3 That, either in measurement or in computation, each standard error of any estimate from the results of the sample
item be weighted in inverse proportion to its probability of itself, and
selection. This latter criterion should not be departed from; for 6.6.3 Sources of possible bias in the sampling procedure or
example, equal weights should not be used when the probabili- in the estimating formulas, together with data pertaining to the
ties of selection are unequal, unless calculations show that possible magnitudes of the biases and their effects on the uses
biases introduced thereby will not impair the usefulness of the of the data.
results. 6.7 The development of a good sampling plan will usually
6.2 To meet the requirements of 6.1.1 and 6.1.2, the take place in steps, such as:
sampling plan must describe the sampling units and how they 6.7.1 A statement of the problem for which an estimate is
are to be selected. To meet requirements of 5.1.1, the sampling necessary,
plan must specify that the selection will be made objectively at 6.7.2 Collection of information about relevant properties of
random. To achieve random selection, a table of random the material to be sampled (averages, components of variance,
numbers or a sequence of random numbers generated by a etc.),
random number generator may be used. Random number 6.7.3 Consideration of a number of possible types of sam-
generation is commonly available in commercial software. For pling plans, with comparisons of overall costs, precisions, and
a discussion of sample size related to specified precision, see difficulties,
Practice E122. 6.7.4 An evaluation of the possible plans, in terms of cost of
sampling and testing, delay, supervisory time, inconvenience,
6.3 In meeting the requirements of 6.1.3, carefully state the
6.7.5 Selection of a plan from among the various possible
purposes served by sampling, lest a relatively unimportant aim
plans, and
overbalance a more important one. For example, estimates of
6.7.6 Reconsideration of all the preceding steps.
the overall average quality of a stock of items may be less
important than the rational disposition of subgroups of the 7. Selection of Sample
stock of inferior quality. In this case the method of using
subsamples of equal size drawn from each subgroup is more 7.1 Calculation of the margin of error or the risk in the use
efficient, although at some expense to the efficiency of the of the results of samples is possible only if the selection of the
estimate of the overall average quality. Similarly, in acceptance items for test is made at random. This is true whether the
inspection, samples of equal size drawn from lots that vary procedure is stratified or unstratified.
widely in size serve primarily to provide consistent judgment 7.2 For a method of sampling to be random it must satisfy
with respect to each lot, and secondarily to provide an estimate statistical tests, the most common of which are the “run tests”
of the process average. Where the estimate of the overall and “control charts,” and certain other special statistical tests.
average of a number of lots is the important objective, samples Randomness is obtained by positive action; a random selection
proportional to the sizes of the subgroups will usually yield an is not merely a haphazard selection, nor one declared to be
efficient estimate. For other possible criteria, sizes intermediate without bias. Selection by the proper use of a standard table of
between equal and proportional sampling from the subgroups random numbers is acceptable as random. It is possible and
will be appropriate. feasible to adapt the use of random numbers to the laboratory,
6.4 It is not easy to describe in a few words the many sorts to the field, and to the factory.
of plans that will meet the requirements of 6.1.2 (see Guide 7.3 Mechanical randomizing devices are sometimes used,
E1402). Nor is it easy to describe how these plans differ from but no device is acceptable as random in the absence of
those that do not satisfy the requirement. Many standard thorough tests. The difficulties in attaining randomness are
techniques, such as pure random unstratified sampling, random greater than generally known. Thus, special randomizing
stratified sampling, and sampling with probabilities in propor- devices intended for the production of random numbers have

2
 E105 − 16
often failed to give satisfactory results until adjusted and sample the actual material, groups of items, such as dirt
retested with perseverance. However, mechanical selection is particles, are obtained in increments.
still usually preferable to a judgment-selection. 8.1.3 It can be difficult to apply the basic principle that
7.4 Some other methods of sampling should be mentioned every portion of the population has a specified non-zero
that do not meet the requirements of randomness. For example, probability of being in the sample. This principle becomes
one may declare that a lot of item is “thoroughly mixed,” and impossible to apply when some units are inaccessible, such as
hence that any portion, even the top layer, would give every in odd-shaped containers or cargo holds. Trying to get material
item an equal chance of selection. In the absence of elaborate near the side or bottom of a container can disturb the matter
steps to mix the product, followed by careful tests for nearby. Denser or smaller particles might be near the bottom.
randomness, such assumptions are risky, as they often lead to Similar difficulties of access can affect sampling of discrete
wrong results. units.
8.1.4 Considerations outside the usual sampling sphere
7.5 Again, another common practice is to take a systematic become important. Material may change chemically when
sample consisting of every kth item. Even if the first item is exposed to different pressure, to light, or to the atmosphere.
selected at random, this type of sample, although random, is
8.2 In general, sampling variation can be reduced by taking
actually a sample of only one of the k possible sampling units
smaller increments, taking more increments, reducing the
that can be formed with an interval of k. Hence, in the absence
particle size of solid material before sampling, and mixing the
of knowledge concerning the order of the material, such a
material before sampling. Use a sampling tool that will not
sample does not permit a valid calculation of the standard error.
under- or over-represent certain type of particles. Rounded
Moreover, it does not yield a comparison of the variances
scoops, for example, will under-represent particles near the
between and within groups of units, statistical information that
bottom. Carryover of material can happen when the sampling
might indicate the direction of change toward a more efficient
tools or connectors are not cleaned between sampling events.
sampling plan.
For discussion of particular methods for sampling different
7.6 However, the use of 10 independent random starts kinds of bulk materials, see Practice E300.
between 1 and 10 k, together with every 10 kth unit thereafter, 8.3 As experience is acquired, the sample can be increased
to form 10 independent systematic subsamples does permit a or decreased to meet the requirements more exactly and more
valid calculation of the standard error, together with some economically. In any case, a valid estimate can be made of the
information on the variances between and within groups of precision provided by any probability sample that was selected,
units. based on an examination of the sample itself. In this
7.7 The foregoing paragraphs do not mean that nonrandom connection, random fluctuations that arise from the measure-
and judgment sampling are of no value. A preliminary judg- ment process must be considered and appropriate allowance
ment sample, for example, may provide useful information for made, if necessary.
the efficient design of a probability sampling plan. Again, if the 8.4 Because of the physical nature, condition, or location of
material being inspected is known to vary but little, a “grab” the material at the time of intended sampling, selection of the
sample will be helpful in assessing the level of the character- units specified in a proposed sampling plan may not be
istic concerned. feasible, physically or economically. No matter how sound a
7.8 It also should be noted that judgment plays an important given sampling plan is in a statistical sense, it is not suitable if
role in the design of a probability sampling plan. For example, the cost involved is prohibitive or if the work required is so
it may be used to assess costs, to estimate spreads and likely strenuous that it leads to short cuts or subterfuge by those
values of variances; also definitions of strata. In the actual responsible for the sampling.
probability sample, however, judgment is not used in the 9. Planning for Sampling
selection of the individual items of the sample, nor in making
the inferences, nor in calculating the risks of decisions based 9.1 Different problems or difficulties are encountered with
wholly on the sample of succession of samples. various kinds of materials, and they require specific solutions
for individual cases. Some general features of solutions to
8. Sampling of Bulk Materials common difficulties are as follows:
9.1.1 Lack of specific information on the pertinent statistical
8.1 Sampling of a bulk material involves some similar and characteristics of the class of material to be sampled may
some different principles from probability sampling of discrete sometimes be overcome to a satisfactory degree, without
units. excessive cost or delay, by investigation and utilization of
8.1.1 A sample from the population consists of increments, existing, apparently unrelated data and general information.
not items that can be individually identified. Sample size refers 9.1.2 The cost of a sampling plan is not confined to the
to weight or volume or the number of increments rather than direct monetary costs of sampling and testing. Plans that secure
the number of items. greater simplicity, convenience, or speed at the expense of
8.1.2 Forms of systematic or stratified sampling may still be higher direct costs sometimes have lower total costs and may
used to subdivide the population, but only in a limited sense. then be appropriately adopted.
For example, one may stratify an area of land according to 9.1.3 Random error can sometimes be reduced by proper
ground slope and wind direction. Still, once one starts to stratification. Where physical difficulties are encountered in

3
 E105 − 16
stratified sampling, the statistician requires the cooperation of provided in envelopes for use as needed. In the selection of
the engineer for possible solutions; in any case, the knowledge material from boxes, templates with random cutouts can be
and cooperation of the engineer will be helpful in choosing the used. Units difficult to move in warehouses may be divided
nature and extent of stratification. into rows or stacks or other appropriate subgroups; the
9.1.4 Economic reduction in the variance of the ultimate subgroups, and the units within subgroups, that are to be drawn
sampling unit is sometimes possible, as by a change in size or into the sample can then be determined by the use of random
shape, or by a choice of units that cut across natural strata. numbers. A general rule is that where the use of tables of
9.1.5 Inability to obtain economically the desired sampling random numbers appears cumbersome or costly, there can
units from a lot of material in place is frequently a major usually be found a reformulation of the sampling plan that will
stumbling block in the actual sampling of such material. For minimize the cost without sacrificing the probabilistic nature of
such units to become accessible, the material must be handled the desired estimate.
or moved. Since movement (transportation) is usually neces- 9.1.8 The sampling devices that are used in any given place
sary at some stage in the utilization of the material, consider- can affect enormously the accessibility of the ultimate sam-
ation should be given to the possibility of drawing the sample pling units specified by the sampling plan, and therefore the
at this time. possibility of attaining randomness, and proportionality within
9.1.6 Certain forms of transportation of some classes of bulk
strata. The expenditure of considerable effort is frequently
materials sometimes effect a mixing of the elementary particles
warranted in the development of superior devices. As statistical
of the material, and sometimes a segregation. The sampling
and engineering factors are mutually interacting throughout the
plan may often be modified to take advantage of this mixing or
design of an efficient probability sampling plan, close coop-
segregation. Sometimes a modification in the transporting
eration is necessary between specialists in the two fields. It is
system will emphasize such a change, so that a modified
possible, of course, that adequate specialized knowledge of
sampling plan will permit still more economical sampling.
both fields may be combined in one person.
9.1.7 Selection by use of random numbers need not be more
onerous or costly than hit-or-miss methods of sample selection,
provided the sampling plan is thoughtfully formulated. For 10. Keywords
example, where the actual use of random number tables is 10.1 bulk sample; probability sample; random sample; sam-
difficult, random numbers may be selected in advance and pling plan; stratification; systematic sample

ASTM International takes no position respecting the validity of any patent rights asserted in connection with any item mentioned
in this standard. Users of this standard are expressly advised that determination of the validity of any such patent rights, and the risk
of infringement of such rights, are entirely their own responsibility.

This standard is subject to revision at any time by the responsible technical committee and must be reviewed every five years and
if not revised, either reapproved or withdrawn. Your comments are invited either for revision of this standard or for additional standards
and should be addressed to ASTM International Headquarters. Your comments will receive careful consideration at a meeting of the
responsible technical committee, which you may attend. If you feel that your comments have not received a fair hearing you should
make your views known to the ASTM Committee on Standards, at the address shown below.

This standard is copyrighted by ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959,
United States. Individual reprints (single or multiple copies) of this standard may be obtained by contacting ASTM at the above
address or at 610-832-9585 (phone), 610-832-9555 (fax), or service@astm.org (e-mail); or through the ASTM website
(www.astm.org). Permission rights to photocopy the standard may also be secured from the Copyright Clearance Center, 222
Rosewood Drive, Danvers, MA 01923, Tel: (978) 646-2600; http://www.copyright.com/