UGC NET Questions (Paper – 1)

Statement D is wrong because probability sampling typically requires a sampling frame or some representation of the population to ensure known selection probabilities. Statements A, B, C and E are correct: the sampling frame is the list from which samples are drawn, its quality affects representativeness, outdated frames cause coverage errors and an ideal frame covers the entire target population. Therefore, the only wrong statement is D.

Option A:
Option A is incorrect because it also labels A, which correctly defines the sampling frame, as wrong along with D. This misrepresents a basic concept in sampling.

Option B:
Option B is wrong because it groups C with D as wrong, even though C accurately states that outdated frames cause coverage errors. Coverage error is a standard threat to representativeness.

Option C:
Option C is correct because it isolates D as the sole statement inconsistent with probability sampling principles and implicitly accepts A, B, C and E as true.

Option D:
Option D is incorrect because it treats B and E as wrong along with D, despite both B and E properly describing the relationship between frame quality and sample representativeness.

Stratified random sampling involves dividing the population into relatively homogeneous strata based on characteristics such as gender, region or discipline and then drawing random samples from each stratum. This ensures that important subgroups are adequately represented in the final sample. It can increase the precision of estimates, especially when subgroup differences are substantial. Because the stem mentions dividing the population into homogeneous subgroups and sampling from each, it describes stratified random sampling.

Option A:
Stratified random sampling uses random selection within each stratum, preserving probabilistic principles while controlling subgroup representation. This approach leads to a sample that mirrors the composition of the population with respect to chosen stratification variables. These features align exactly with the process outlined in the stem, making stratified random the correct answer.

Option B:
Simple random sampling selects units from the population without any prior subdivision, giving each element an equal chance of being chosen, but it does not guarantee representation of specific subgroups. Since the question specifies dividing the population into subgroups first, simple random is not the best match.

Option C:
Systematic sampling selects every kth element from an ordered list after a random start, without necessarily dividing the population into strata. Although it is a probability method, it does not follow the subgroup-based approach described in the stem. Therefore, systematic is not appropriate here.

Option D:
Cluster sampling divides the population into clusters, often based on natural groupings such as schools or villages, and then selects some clusters, usually sampling all or many elements within chosen clusters. Clusters are not necessarily homogeneous, and the method focuses on groups rather than ensuring representation from all subgroups. Thus, cluster sampling does not fit the description given.

In this question statements A, B and C appropriately describe the concepts of population, sample and probability sampling. Statement A is true because the population refers to the total set of units about which the researcher wants to generalise. Statement B is correct in defining a sample as a subset of that population, and statement C accurately reflects that in probability sampling every element has a known non-zero chance of selection. Statements D and E are wrong: D incorrectly labels convenience sampling as a rigorous probability technique, and E wrongly suggests that a large sample size alone guarantees representativeness regardless of sampling method, so the wrong statements are exactly D and E.

Option A:
Option A brings together statements D and E, both of which misrepresent sampling concepts. It recognises that convenience sampling is a non-probability method and that large sample size without appropriate sampling design cannot ensure representativeness. Because it neither omits any wrong statements nor includes any correct ones, this combination is the correct answer.

Option B:
Option B treats statement D as the only wrong statement and thus ignores the error contained in statement E. While it is true that calling convenience sampling a rigorous probability method is incorrect, it is equally incorrect to claim that sample size alone guarantees representativeness. Since this option fails to capture all the wrong statements, it cannot be accepted.

Option C:
Option C identifies only statement E as wrong and overlooks statement D. It is right to challenge E because representativeness depends on the sampling technique and frame, not just on the size. However, by failing to mark D as wrong, this option does not fully answer the question.

Option D:
Option D erroneously includes statement C among the wrong statements. Statement C correctly defines probability sampling as giving every element a known non-zero chance of selection. By classifying a correct statement as incorrect, this combination becomes logically inconsistent and therefore cannot be the correct answer.

In any research study, the population is the complete set of units about which the researcher wants to draw conclusions. Because studying all units is often impractical, a smaller group called a sample is selected from this population. The sample should represent the population so that inferences from sample data can be generalized appropriately. This distinction is fundamental to sampling theory.

Option A:
Option A incorrectly claims that population is always smaller than sample, which is logically impossible because a sample is drawn from the population. Samples are subsets and therefore cannot exceed the population in size.

Option B:
Option B clearly describes population as the full group of interest and the sample as a subset selected for actual study. This matches standard methodological definitions and underscores the logic of generalizing from sample to population.

Option C:
Option C states that sample and population are identical, which contradicts the very meaning of sampling. If they were identical, there would be no need to use sampling techniques.

Option D:
Option D incorrectly restricts the term “sample” to qualitative data and “population” to quantitative data. In reality, both populations and samples can involve any type of data; the distinction is based on whether we consider all units or only a subset.

Probability sampling is defined by the use of random selection mechanisms that give each member of the population a known, non-zero chance of inclusion. This property allows researchers to estimate sampling error and to generalise findings from the sample to the population with quantifiable confidence. Techniques such as simple random, stratified and cluster sampling fall under this category. Because the stem explicitly mentions a known non-zero chance of selection for every element, it is describing probability sampling.

Option A:
Quota sampling is a non-probability method where the researcher fills pre-specified quotas for certain subgroups, often using convenience within each quota. Individual elements do not have known chances of selection, and probability theory cannot be applied to estimate sampling error. Thus, quota sampling does not satisfy the conditions stated in the stem.

Option B:
Purposive sampling involves selecting cases that are judged to be information-rich or particularly relevant to the research question, based on the researcher’s judgement. It does not assign known probabilities of selection to all population members, so it is not probability sampling. Therefore, purposive sampling cannot be the correct option.

Option C:
Convenience sampling selects units that are easiest to access, such as volunteers or people who are readily available. This method ignores the requirement that every member of the population have a known non-zero chance of selection, and it often yields biased samples. Consequently, convenience sampling does not meet the criteria in the question.

Option D:
Probability sampling uses objective randomisation procedures, such as random number tables or computer-generated random lists, to ensure fairness in selection. Because each element’s chance of being chosen is known, researchers can make statistically grounded inferences to the population. This matches exactly what the stem describes, making probability sampling the correct answer.

Inductive generalisation infers a probable conclusion about a population based on evidence from a sample. A large random sample in which 85% have internet access supports the claim that most households in the population likely have access. However, it does not justify an exact percentage or absolute statements about all households. The word “probably” reflects the probabilistic nature of the inference correctly.

Option A:
Option A overstates the precision by claiming that the population must have exactly the same percentage as the sample, ignoring sampling error.

Option B:
Option B cautiously states that most households have access, which aligns with the sample finding while acknowledging some uncertainty.

Option C:
Option C falsely claims that no household lacks internet, which directly conflicts with the 15% of sampled households without access.

Option D:
Option D attributes a legal or regulatory requirement for access, which goes far beyond the descriptive data.

A sample is a smaller group of units selected from a larger population so that information collected from it can be used to infer characteristics of the whole. Sampling makes research feasible when studying every member of the population is impractical or impossible. The quality of the sample strongly affects the accuracy of conclusions drawn from the study. Since the stem speaks of a subset drawn from the target population for making inferences, the correct term is sample.

Option A:
Population refers to the entire group of individuals, objects or events that share specified characteristics and about which the researcher wants to draw conclusions. It is not a subset but the full set of interest. Because the stem clearly mentions a subset drawn from the population, population cannot complete the statement.

Option B:
A sample is selected using some sampling design so that it reflects the key characteristics of the population as closely as possible. Data are collected from the sample, and statistical techniques are then used to estimate population parameters. This use of a subset for inference corresponds exactly to the description in the question, confirming sample as the correct option.

Option C:
Parameter is a numerical characteristic of a population, such as the true mean or proportion. It is not itself a group of units but a value that describes the population. The stem is looking for the name of the subset of units, so parameter is not appropriate.

Option D:
Universe is another term sometimes used interchangeably with population to refer to the total collection of elements under study. It does not denote a subset selected for research. Hence, universe cannot be the right completion of the stem.

Nonresponse bias arises when certain types of individuals are less likely to participate or return questionnaires. If the nonrespondents differ in important ways from respondents on variables of interest, estimates based on the responding sample will be distorted even when probability sampling was used.

Option A:
This option describes an ideal situation of full response, in which nonresponse bias would not occur because everyone participates.

Option B:
This option correctly explains that bias occurs when nonrespondents are systematically different from respondents; their absence skews the results and undermines representativeness.

Option C:
Using simple random sampling helps avoid selection bias at the sampling stage but does not remove the possibility that some selected individuals will refuse or fail to respond.

Option D:
Whether a questionnaire has open or closed items affects the type of data collected, not whether responders and nonresponders differ systematically.

A sampling frame operationalizes the target population by listing units that can potentially be included in the sample. Examples include voter lists, school registers or membership databases. The quality of the frame affects representativeness because omissions or duplications can bias selection. Thus, it is the practical source list from which sample units are chosen.

Option A:
Option A defines the population, which is the broader group of interest, not necessarily the actual list used for sampling.

Option B:
Option B is correct as it highlights the frame as the list or device from which sampling occurs. It connects the conceptual population to the operational step of sample selection.

Option C:
Option C refers to the achieved sample or respondents, which may be smaller than the selected sample due to nonresponse. This is not the sampling frame.

Option D:
Option D describes the questionnaire, which is a data collection tool, not the list of units.

Sampling error is the random difference between values obtained from a sample and the true values in the population. It occurs because the sample, even if selected properly, represents only a subset of the population. As a result, estimates like means or proportions fluctuate from sample to sample. Larger and more representative samples reduce sampling error but cannot eliminate it entirely.

Option A:
Option A describes errors arising from mistakes in coding or tabulation, which are classified as non sampling errors. These can occur even in a census where the whole population is measured.

Option B:
Option B correctly explains sampling error as the natural variation due to observing only a part of the population. It captures the essence of random fluctuation inherent in the sampling process and is consistent with statistical theory.

Option C:
Option C refers to deliberate response distortion or bias, which again falls under non sampling errors, specifically response bias. While serious, it is conceptually distinct from sampling error.

Option D:
Option D deals with faulty instruments, another form of non sampling error because it affects measurement irrespective of whether a sample or census is used.

Probability sampling is characterised by the use of random selection mechanisms that give each element in the population a known, non-zero chance of being included in the sample. This property allows researchers to estimate sampling error and use statistical inference to generalise from the sample to the population. Techniques such as simple random, stratified and systematic sampling fall under this category. Therefore, the sampling procedure described in the stem is correctly called probability sampling.

Option A:
Non-probability sampling does not give every member of the population a known chance of selection, often relying on convenience, judgement or voluntary participation. As a result, sampling error cannot be strictly quantified, and generalisations are more tentative. Hence, non-probability sampling is not the correct completion.

Option B:
Convenience sampling selects whoever is easiest to reach, without any randomisation or known probabilities of selection. It is prone to bias and lacks the formal properties required for probability-based inference, so convenience sampling does not match the stem.

Option C:
In probability sampling, each element’s inclusion likelihood can theoretically be specified, which underpins many statistical tests and confidence interval calculations. This feature aligns exactly with the requirement stated in the question.

Option D:
Judgement sampling, also known as purposive sampling, involves selecting units based on the researcher’s subjective judgement about their relevance or typicality and thus does not provide a known non-zero probability of selection. Therefore, judgement sampling is not appropriate here.

Non-probability sampling refers to techniques in which elements of the population are selected using non-random procedures such as convenience, purposive or quota sampling. In these methods, the exact probability of any given element being included in the sample is unknown. This limits the ability to generalise findings using statistical inference. Because the stem explicitly states that selection probabilities are unknown, non-probability sampling is the correct answer.

Option A:
Probability sampling, by contrast, ensures that each element of the population has a known, non-zero chance of selection, often through random processes. This property enables estimation of sampling error and supports strong generalisation to the population. Since the question mentions unknown probabilities, probability sampling cannot complete the stem correctly.

Option B:
Non-probability sampling is often used in exploratory or qualitative studies, or when constructing sampling frames is difficult or impossible. Although it may introduce bias, it can be practical and informative in certain contexts. These characteristics fit the description provided in the stem, confirming non-probability as the appropriate term.

Option C:
Systematic sampling, a form of probability sampling, selects every kth element from an ordered list after a random start, so individual selection probabilities are known. Thus, systematic sampling does not align with the condition of unknown probabilities.

Option D:
Stratified sampling, another probability method, divides the population into homogeneous groups and draws random samples from each stratum. Here again the chance of selection is known and can be calculated, which contradicts the stem. Therefore, stratified sampling is not the right completion.

Random sampling gives each unit in the population a known, often equal, chance of being selected, which makes the selection process impartial. This greatly reduces systematic bias in the composition of the sample because personal judgment and convenience are minimised. While it does not completely eliminate all error, it helps ensure that differences between the sample and population are mainly due to chance. Therefore, the correct effect on selection bias is that random sampling tends to reduce it.

Option A:
If random sampling were to increase bias, it would defeat its main purpose in scientific research. The entire rationale for using random methods is to minimise systematic distortions in the sample. Thus, the option suggesting an increase in bias is conceptually incorrect.

Option B:
Random sampling reduces the likelihood that any particular type of individual is overrepresented or underrepresented for systematic reasons. It supports the use of probability theory to estimate sampling error and build confidence intervals. These features correspond directly to the idea of reducing selection bias as stated in the question.

Option C:
Creating bias would mean deliberately skewing the sample in favour of certain subgroups, which is characteristic of convenience or judgement sampling, not random sampling. Therefore, the suggestion that random procedures create bias is misleading.

Option D:
Ignoring bias does nothing to protect the quality of inferences and would be a sign of poor methodological practice. Random sampling is chosen precisely because it addresses concerns about bias rather than ignoring them, so this option does not fit the role described in the stem.

Non-response bias occurs when the characteristics of non-respondents differ in important ways from those of respondents, leading to distorted estimates of population parameters. For example, if less satisfied individuals are less likely to answer a satisfaction survey, the results will be overly positive. This bias persists even with large sample sizes if non-response is systematic. Therefore, the bias described in the stem is correctly termed non-response bias.

Option A:
Sampling bias refers more broadly to systematic errors in the way the sample is selected, such as excluding certain groups entirely, but it does not specifically focus on the difference between respondents and non-respondents. Non-response is one particular source of sampling error, so sampling bias is too general here.

Option B:
Measurement bias arises when instruments or questions systematically overestimate or underestimate the true values, for instance due to poorly worded items or faulty devices. It is unrelated to whether people choose to respond or not. Hence, measurement bias is not the appropriate completion.

Option C:
Non-response bias undermines the representativeness of survey findings because missing data are not random but correlated with variables of interest. Strategies such as follow-up contacts or weighting adjustments are often used to reduce its impact. These aspects fit exactly with the description in the question.

Option D:
Response-set bias involves patterns such as acquiescence or social desirability in the way respondents answer questions, affecting the quality of obtained responses rather than whether people respond at all. Therefore, response-set bias does not match the situation described in the stem.

Inductive generalisations draw probable, not certain, conclusions from sample data. If the 100 commuters were chosen randomly and the sample reflects the diversity of city commuters, the observed 70% support provides reasonably strong evidence that most commuters prefer the metro. However, sampling error and bias remain possible, so the conclusion is not guaranteed. The cautious phrase “probably reasonable” accurately reflects the strength of this inductive inference.

Option A:
Option A treats the inductive conclusion as certain, ignoring the role of sampling variability and possible bias. Even a good sample cannot guarantee exact population proportions.

Option B:
Option B recognises that a well chosen sample can justify a probable claim about the population while still leaving room for uncertainty. This matches standard statistical reasoning.

Option C:
Option C dismisses all inferences from a sample of 100, which is unjustified; while larger samples are better, 100 can still yield useful information when properly drawn.

Option D:
Option D confuses the target group, since commuters are precisely the relevant population for this question; interviewing them is appropriate, not a flaw.