The Oxford Advanced Learners Dictionary of Current English defines data as the fact if information especially when examined and used to find out things to make decisions or information. The importance of data gathering in legal research is many. It enables the researcher to secure accurate information on a research topic. In this regard, collection of data enables to understand the object of study, the events and the phenomenon in the research, to know the extent and limitations on the information available on request topic.
Most research studies are based on samples. When a small group is selected as representative of the whole, it is known as sample method. Sampling can be defined as the method or the technique consisting of selection for the study of the so called part or the portion or the sample, with a view to draw conclusions or solutions about the universe or the population. In order to answer the research questions, it is doubtful that the researcher should be able to collect data from all cases.
Thus, there is a need to select a sample. The entire set of cases from which researcher sample is drawn in called the population. Since, researchers have neither the time nor the resources to analyze the entire population so they apply sampling technique to reduce the number of cases. This law comes from the mathematical theory of probability. The Purpose of Sampling: In some types of research the target population might be as broad as all humans, but in other types of research the target population might be a smaller group such as teenagers, preschool children or people who misuse drugs.
It is more or less impossible to study every single person in a target population so psychologists select a sample or sub-group of the population that is likely to be representative of the target population we are interested in. This is important because we want to generalize from the sample to target the population. The more representative the sample, the more confident the researcher can be that the results can be generalized to the target population.
One of the problems that can occur when selecting a sample from a target population is sampling bias. Sampling bias refers to situations where the sample does not reflect the characteristics of the target population. The population or universe embodies the entire group of units which is the centre of the study. Thus, the population could consist of all the persons in the country, or those in a particular topographical position, or a special cultural or economic group, depending on the rationale and exposure of the study.
Thus, it is a total set of elements persons or objects that share some common features defined by the sampling criterion established by the researcher. A sample is the group of units who took part in research. Generalisability refers to the degree to which we can correlate the findings of our research to the target population we are concerned. This population is a split or subset of the target population and is also known as the study population.
It is from the accessible population that researchers draw their samples. Sample Group or Sampling. It is the most bias thing in the universe from which data is to be collected. For example, in a study proposed for assessing the violation of human rights among hand-rickshaw pullers in the city 8. Herein, the universe will be the entire body of rickshaw pullers in Kolkata. In some studies more than one sample is drawn out of the universe for making a sound research.
Size of the sample is the total number of sampling units that the researcher will include in the sample. The size of the sample should not be vast as the purpose of studying the sample and not the universe will be lost. Similarly, the sample size cannot be too small either for it will not adequately represent the universe.
Such a sample is called a biased sample. It is pertinent for the researcher to be aware and make sure that his samples are not biased, to avoid error in sampling. It is customary for the researcher to mention the research loopholes that led to the result.
While sampling errors can be predicted quite precisely as they can be calculated, the non-sampling errors can only be guessed or assumed by the researcher. Sampling errors arise due to wrong selection of samples and can be avoided is the researcher is cautious in choosing the sampling technique.
Non-sampling errors arise in the pre or post sampling process of a research. Some common sampling methods are simple random sampling,stratified sampling, cluster sampling, quota or judgment. Different sampling methods may use different estimators. For example, the formula for computing a mean score with a simple random sample is different from the formula for computing a mean score with a stratified sample.
Similarly, the formula for the standard error may vary from one sampling method to the next. The best sample design is dependent upon survey objectives and on survey resources. For example, a researcher might select the most economical design that gives a required level of accuracy. Or, if the resources are limited, a researcher might select the design that gives the greatest accuracy without going over financial plan.
Characteristics of a good Sample Design: In a field study due to time constraint and finance involved, generally, only a section of the population is considered. These respondents are identified as the sample and are representative of the general population or universe. A sample design is a predetermined plan for getting a sample from a population.
It refers to the method or the process for obtaining a sample from a given population. This sample is required to match all the features of the entire population. If the sample Sampling error refers to the difference that may result from judging all on the basis of a small number. Summary of relative efficiency of systematic sampling designs vs simple random sampling designs for the freshwater mussel population.
Abbreviations and design descriptors as in Tables 1 and 2. Sampling- No. Mean RBV is shown for each design for the clustered real population A and the random simulated population B. Abbreviations and design descriptors are as in Table 1. Bias was calculated using estimates of variance appro- priate for SYS equation 5.
The dashed horizontal line at 0 represents no bias. In those instances, mean RB was simulated populations, but deviations from ex- 0.
The systematic formulae, number of random starts beyond 3 did not re- however, did result in substantially lower cov- sult in any additional decrease in bias Fig.
RBV estimates for the simulated population did decrease as number of random Quantifying degree of spatial clustering starts increased Fig. Positive bias indicates that the spa- CIs calculated using SRS formulae deviated tial distribution of a particular sample is likely from the expected coverage 0. Mean confidence interval CI coverage for each group of sampling designs used to simulate sampling a spatially random simulated population or a clustered real population see text for details.
Abbre- viations and design descriptors are as in Tables 1 and 2. Number of designs indicates how many designs were examined for each group specified. Equations used for finding values in each column are noted in column headings.
Blank spaces indicate equations 4, 5, and 6 could not be used because the designs had only 1 RS see text for details. Confidence interval coverage Simulated population Real population Sampling- unit size No. When sam- domness. The probability of Type I error prob- pling the real population, the Hopkins Statistic ability of classifying a random population as was negatively biased when a 0.
This over-sensitiv- unit was used and positively biased when a 1- ity did not appear to decrease when sampling m2 sampling unit was used, and relative bias of effort was increased, the number of random the Hopkins Statistic decreased slightly as sam- starts in SYS designs was increased, or an op- pling effort increased Fig. The Type the simulated population, the Hopkins Statistic II error probability of classifying the clustered was positively biased at low levels of sampling real population as spatially random showed effort but had little or no bias as effort increased.
SYS designs rates. MMSE was lowest sample-unit sizes. Moreover, the subjectivity signs using 0. Our empirical re- of error in these results, but one that we could sults indicated that the Hopkins Statistic was not quantify. Bias in 3 measures of spatial clustering as a function of sampling effort using simple random sam- pling SRS and systematic sampling SYS designs with different sampling-unit sizes 0.
Dashed horizontal lines at 0 represent no bias. Discussion mation for estimating abundance, quantifying spatial clustering, and predicting spatial distri- Optimal design bution. Comparisons of RBV estimates indicated SYS designs with 2 or 3 random starts using that using 3 random starts provides more ac- 0. Type I error rate is the probability that a random distribution is incorrectly classified as clustered and was calculated using samples from the random simulated population.
Type II error rate is the probability that a clustered population is incorrectly classified as random if a 5 0. Type II error rate was calculated using samples from the clustered real population. Groups of designs are identical to those in Table 4. However, using. These tered population. However, they also found that same design recommendations also provide the high per unit setup costs would reduce the com- most accurate CI coverage.
The disparities in CI parative advantage of smaller units. Our results corroborate the utility of SYS for Size of sampling unit spatial prediction and to identify optimal de- We recommend 0. Thus, more sampling units designs. All SYS de- overwhelming. A survey with more sampling signs had 3 random starts and were classified by the units will provide better spatial coverage of the distance between sampling units across the stream study area, a larger sample size, and more pre- DAC and the distance between units along the stream cise estimates than a survey with fewer sam- DAL.
Our comparisons among designs predictions taken from each of 50 sampling realiza- differing only in sampling-unit size confirmed tions for each design. Values within a column with the previous studies indicating that surveys using same letter are not significantly different p. SMITH [Volume 24 signs in terms of multiple starts and distance large animal survey in which animal locations between units when applied to freshwater mus- are the sampling units In contrast, we counted sel surveys.
The calcu- based on SYS for spatial interpolation to identify lations, which involve distances to nearest locations of mussel concentration in a large site neighbors, may be affected by some distances 18, m2 and overlaid the predicted spatial between individual mussels that are within the distribution with the locations of potential im- same sampling unit.
These relatively small dis- pacts from bridge replacement. Such analyses of tances would not be present if the sampling spatial distribution can be used to assess the units were the individuals themselves rather likelihood of negative impacts and to protect than the sampling units within which individ- critical populations. Our findings verify that a uals were found. Fac- spatial distribution. SYS lends itself well to in- tors such as prevalence, spatial distribution of terpolation methods such as kriging Thompson the sample population, and idiosyncrasies of the because it satisfies the uniformity condi- area to be sampled must be taken into consid- tion needed.
As a practical consideration, SYS eration. In addition, almost all sampling studies with multiple random starts is easier to imple- are limited by cost, labor, and, in many cases, ment than SRS, and it retains a random com- weather constraints. In light of these issues, SYS ponent that allows valid statistical inference.
The recommendations we have made across the current should be equal to or less for sampling a clustered freshwater mussel pop- than the distance between units along the cur- ulation e.
If information on directional varia- may not fit the goals of studies involving other tion within the study area is known, sampling species. However, regardless of species, it makes units can be set closer together in the direction sense to place sampling units closer together in of more variation. Variation was greater across the direction of more variability. If no informa- the current than along the current in the fresh- tion about directional variation is known, plac- water mussel population we examined, and we ing sampling units equal distances apart in all expect that spatial pattern to hold generally for directions is appropriate for a wide variety of freshwater mussels in lotic environments.
We think that almost all sam- pling studies could benefit greatly from doing a Spatial clustering relatively small pilot study ahead of time to ob- tain estimates of variability.
Information gained The erratic bias of the Hopkins Statistic raises from a pilot study is often invaluable in design- some interesting questions about using nearest- ing an appropriate sampling scheme.
Cressie emphasized David Weller, and Priscilla Young assisted in the extreme sensitivity of the Hopkins statistic data collection and data management. We thank when sampling a clustered population, but he Eric Smith for a helpful review prior to submis- did not discuss its accuracy when sampling a sion. Two anonymous referees and David Stray- spatially random population. Cressie dis- er provided detailed critiques of earlier versions cussed the effectiveness of using the Hopkins of our manuscript.
Methods of estimating the population sample surveys where the goal is to sample of insects in a field. Biometrika — We discuss briefly the scope of statistics in some of the following fields. Statistics in industry: industry makes use of statistics at several places such as administration, planning, production, growth and development.
In many industries statistical quality control division is separately operating. Moreover purchased goods or semi finished goods are inspected using acceptance sampling plans of various types.
Now-a-days, ISO makes use of statistics to the large extent. Newly installed machinery is tested for its performance using statistical methods. Sampling is required to be used because of its several advantages.
Web: commercedigest. Statistics and economics: In the field of economics, huge amount of data are needed to be processed and Interpreted.
Statistics is very much useful in this field. In order to collect data, various statistical methods of investigations are used. Many a times questionnaire was drafted. A proper representative of a group is selected using sampling methods. Statistical methods are used in this activity to get reliable results. Estimation of national income, per capita income, poverty line, industrial production etc.
An index number developed in statistics is used every now and then in economics. Statistics and management: Most of the managerial functions make use of statistics.
For efficient working of various sections of management such as sales, production, marketing and statistical methods are used.
Different statistical tools such as forecasting, test of significance, index numbers, time series analysis, statistical quality control, estimation play vital role in management activities. Apart from this, various optimisation techniques known as linear programming, transportation techniques, job assignment problem, sequencing, CPM and PERT, replacement problem, inventory control are also useful.
Research in social science need questionnaire. Further analysis is required to be done using statistical tools. In social sciences we need to test association between two variables such as education and criminality, education and marriage adjustment score, gender and education, richness and criminality etc. We use Techniques of sampling several times in everyday life. For example, while purchasing food grains we inspect only handful of grains and draw conclusion about the whole sack.
Sampling is a well accepted means of collecting information. Moreover it is believed to be scientific and always in your procedure of selecting items. Sampling plays an important role in statistical inference. Population: In the technical language of statistics the word population is used in somewhat for wider sense.
It does not mean only a human population. For example i in the study of industrial development, of the industries under consideration is the population. Ii In the study of socio economic conditions of a particular village, all families or houses in the village will be a population. Iii In the study of agricultural yield, all the cultivated farms together will be population. Definition: An aggregate of object or individuals under study is called population or universe.
Population may contain finite or infinite elements.
0コメント