UGC NET Questions (Paper – 1)

A ratio scale has all the properties of an interval scale—equal intervals between values—and, in addition, a true zero point that indicates the complete absence of the measured attribute. This allows statements such as one value being twice or half of another to be meaningful. Examples include height, weight and reaction time. Because the stem highlights equal intervals, a true zero and meaningful ratios, it is describing a ratio scale.

Option A:
Interval scales, such as temperature in Celsius, have equal intervals but lack a true zero point, which makes ratio statements (like “twice as hot”) conceptually problematic. Therefore, interval is not sufficient to capture the “true zero and meaningful ratios” criterion given in the question.

Option B:
Ratio scales permit the full range of mathematical operations, including addition, subtraction, multiplication and division. This rich set of properties enables powerful statistical analyses and interpretable comparisons. These features match the description in the stem, so ratio is the correct completion.

Option C:
Nominal scales classify data into categories without order or quantitative meaning, making arithmetic operations and ratio statements inappropriate. They lack both equal intervals and a true zero, so nominal cannot be the answer.

Option D:
Ordinal scales indicate order among categories but do not assume equal differences between ranks, and they lack a true zero point. While they allow ranking, they do not support meaningful ratio comparisons, making ordinal an incorrect option for the stem’s description.

A normal distribution is a theoretical continuous probability distribution characterised by perfect symmetry about its mean, with the mean, median and mode all equal at the centre. The tails extend infinitely in both directions, and many natural and social phenomena approximate this shape under certain conditions. It plays a central role in inferential statistics because many tests assume normality of errors or variables. Because the stem describes a symmetrical, bell-shaped distribution with coinciding mean, median and mode, normal distribution is the correct term.

Option A:
Skewed distributions are asymmetrical, with one tail longer or fatter than the other, causing the mean, median and mode to separate rather than coincide at the centre. They do not fit the perfectly symmetrical, bell-shaped description given in the stem. Therefore, skewed is not the right answer.

Option B:
Normal distributions are fully described by their mean and standard deviation, and many sampling distributions approach normality under the central limit theorem. This makes them fundamental to constructing confidence intervals and conducting many parametric tests, which aligns with the importance implied in the question.

Option C:
A bimodal distribution has two distinct peaks, representing two modes, and is not necessarily symmetric or bell-shaped in the classic Gaussian sense. As the stem indicates one central peak where the three measures of central tendency coincide, bimodal distribution cannot be the correct completion.

Option D:
A uniform distribution has equal probability across its range, often represented by a rectangular shape rather than a bell. It does not exhibit the central peak and tapering tails characteristic of the normal curve, so uniform is not appropriate here.

Sampling error refers to the natural discrepancy between sample statistics and population parameters that arises because only part of the population is observed. Even with perfectly random sampling and accurate measurement, different samples will yield slightly different estimates. This variability is quantified in terms such as standard error and confidence intervals. Hence, the difference described in the stem is correctly called sampling error.

Option A:
Measurement error occurs when the instruments or procedures used do not yield perfectly accurate values, leading to inaccuracies that are distinct from the randomness of sampling. It is related to how variables are measured rather than to the fact that only a subset is observed. Therefore, measurement error is not the best completion.

Option B:
Systematic error is a consistent bias in measurement or procedure that pushes results in one direction, such as miscalibration or interviewer bias. It does not average out across samples and is not due simply to using a subset. Consequently, systematic error is not what the stem is defining.

Option C:
Sampling error arises even in well designed probability samples and can be estimated and controlled by appropriate sample sizes. It reflects sample-to-sample fluctuation, making it the correct term for the discrepancy between sample and population values mentioned in the question.

Option D:
Non-response error occurs when individuals who fail to respond differ systematically from those who do, introducing bias, but this is a specific non-sampling error, not the inherent fluctuation due to partial observation. Thus, non-response error is not the accurate completion.

A ratio scale possesses all the properties of an interval scale—equal units and meaningful differences—but also includes a true zero point that indicates the absence of the measured attribute. This allows researchers to say that one score is twice or half another in a meaningful way, as in weight or height. Because of these features, ratio scales support the full range of mathematical operations. Thus, the scale described in the stem is correctly called the ratio scale.

Option A:
Ordinal scales allow ranking of cases but do not guarantee equal intervals between ranks, nor do they have a true zero. Numbers indicate order only, not the magnitude of differences, so ordinal scale does not match the properties described in the question.

Option B:
Nominal scales provide categorical labels without quantitative meaning or order and have no concept of interval or zero magnitude. They cannot support ratio statements like “twice as much.” Hence, nominal scale is not appropriate here.

Option C:
Interval scales do have equal units, enabling meaningful comparison of differences, but they lack a true zero point, as seen in Celsius temperature. This absence prevents meaningful ratio interpretations, so interval scale falls short of the description given in the stem.

Option D:
Ratio scales, by including a natural zero, make statements like “this object weighs twice as much as that one” legitimate, which is exactly the property highlighted in the question. This confirms ratio as the correct answer.

A nominal scale assigns symbols or numbers to categories simply to distinguish them, such as coding males as 1 and females as 2, without any inherent ranking. The categories are mutually exclusive and exhaustive, but the numbers have no quantitative meaning. Only equality or difference between categories can be assessed. Therefore, a labelling operation without order is known as nominal scaling.

Option A:
An ordinal scale does imply a meaningful order among categories, such as ranks or levels of agreement, where higher numbers correspond to higher positions. Since the stem specifies that numbers do not imply any order, ordinal scale cannot be correct.

Option B:
Nominal scales are common for variables like religion, occupation or region, where categories are different but not ranked. Statistical operations are limited to counts and mode, which aligns with the purely classificatory function described in the stem.

Option C:
Interval scales have equal units and allow meaningful addition and subtraction but lack a true zero, such as temperature in Celsius. They provide more information than simple labels and thus do not correspond to the non-ordered classification mentioned in the question.

Option D:
Ratio scales possess all the properties of interval scales plus a true zero, enabling meaningful ratio comparisons, such as height or weight. They go far beyond mere labelling, so ratio scale is not the appropriate completion for the stem.

Multiple correlation refers to the degree of association between a single criterion variable and several predictor variables considered simultaneously. It is summarised by the multiple correlation coefficient R, which indicates how well the combined predictors explain variance in the criterion. This is different from simple correlation, which involves only two variables. Because the stem mentions correlation between one variable and a combination of two or more predictors, multiple correlation is the correct term.

Option A:
Partial correlation measures the relationship between two variables while statistically controlling for the effect of one or more additional variables. It does not directly express the combined predictive power of several variables on one criterion as multiple correlation does. Thus, partial is not the appropriate completion.

Option B:
Simple correlation, often denoted by r, captures the linear relationship between two variables at a time. It does not account for the simultaneous effects of multiple predictors on a single outcome. Since the stem explicitly refers to a combination of predictors, simple correlation is not correct.

Option C:
Multiple correlation summarises how well a set of predictors, such as intelligence, study hours and motivation, explain a criterion like academic achievement. A high R value indicates strong combined prediction. This conception aligns closely with the stem’s description, confirming multiple as the right answer.

Option D:
Serial correlation is often used to describe correlations between observations ordered in time, such as successive scores in a time series, and is not directly about combining predictors for a single criterion. Hence, serial is not suitable here.