Quantities in statistics: absolute, relative, average, their application in comparing populations"
Year: 2025
Type: article
Abstract: Quantities in statistics: absolute, relative, average, their application in comparing populations” An article by Rutuja Aher and Sakshi Nigal under guidance of moldoev sir Abstract: The article examines the basic statistical measures of absolute numbers, relative numbers (rates, ratios, proportions), and averages (mean, median, mode). The definitions, calculation, and most importantly, utilization for comparisons across populations is considered. While absolute numbers afford an initial count, they are often inadequate for useful comparisons. Relative numbers provide a means of standardization, allowing for the differences in population size, and produce vastly more accurate information. Averages also provide a summarizing measure of central tendency, but should not be used alone without considering accompanying measures of dispersion. Using these measures appropriately, while keeping in mind their weaknesses, is necessary for sound inference whether comparing populations in select disciplines, including public health, demography, and sociology. Introduction Statistics is an important tool for analyzing, understanding, and interpreting outcomes from both data and populations. When comparing two or more populations, using appropriate statistical quantities is important to prevent misleading conclusions to be drawn. If you only use a comparison of raw counts (absolute counts) populations may differ in size and composition so often can be misleading. Relative percentages, proportions and averages are ways to standardize data and highlight a key trait. and therefore provide more nuanced ways to conduct comparisons. The purpose of this article is to review properties of, application, and limitations for each of absolute, relative and average quantities for application across populations. Content: 1. Absolute Numbers: • Definition: Absolute numbers refer to the overall frequencies or number of events or individuals within the population. They provide information on the aggregate amount of the event that occurred within this population regardless of other variables. • Examples: It may include the number of deaths of a city, the number of patients affected by a specific disease, or the number of students studying in a specific school. • Application: They can be applied to get a better sense of the overall extent of a specific phenomenon that is located in a single population. • Limitations: These estimates cannot be used to compare populations that are different in size. An absolute number of a phenomenon can be higher in one population, and it does not reflect a higher risk or prevalence of the phenomenon in which we are concerned. 2.Relative Numbers: Relative numbers communicate the relationship between two amounts, which facilitate the standardization and therefore more precise comparisons. Relative numbers include, for example, rates, ratios and proportions. * Rates: Which measure the frequency of an event among a population during a certain period, and typically reported as a standardized measure per unit of population size. - To calculate this: (Number of events/ population size) x Constant (e.g. either 1,000 or 100,000) - Examples include birth rates (births per 1,000 population), mortality rates (deaths per 1,000 population) or the incidence rate for a disease (such as new cases per 100,000 population). - The application of rates can facilitate accurate comparisons of frequencies of events regardless of the sizes of the populations. - Clinical significance of using incidence rates are useful in public health to identify population groups that may be at a higher risk of developing a certain disease. This is important to derive intervention strategies that lower future rates of disease in the target high-risk population. * Ratios: Which express the relationship between two different quantities that can often have different units. Quotients: Answer quantity A / Answer quantity B • Example: sex ratio (# of males per 100 females), doctor-to-population ratio (# of doctors per 1,000 population); odds ratio (in case-control studies) • Where we'd find use: Good for comparing relative sizes of different groups or categories within an overall population. • Clinical significance: Case-control studies are often used to identify risk factors associated with a particular disease (or medical condition). Then we calculate odds ratios to quantify association between some exposure factor and development of the disease. An odds ratio > 1 means a positive association and an odds ratio < 1 indicates a negative association between exposure factor and disease. • Proportions: i.e., represent the fraction of the population that has a particular characteristic. * How we would calculate: (Number of individuals with the characteristic / Total population size) * For example: Proportion of vaccinated individuals; proportion of smokers; proportion of individuals living in poverty; - mostly expressed as a percentage * Where we would find use: Good for understanding the distribution of characteristics in a population. * Clinical significance: Prevalence is the proportion of the population who have a condition at a point in time. Prevalence can be used as a study to estimate the total disease burden AND resources needed to manage a condition in a specific cohort. 3. Averages (Measures of Central Tendency):An average conveys a summary measure of a "typical" value in a data set. • Mean: The arithmetic average obtained by summing all values together and dividing by the number of values. * Application: Useful as a summary of data that is normally distributed. * Limitations: Sensitive to outliers (i.e., extreme values). • Median: The middle value after the data is ordered from smallest to largest. * Application: Robust to outliers; preferred for skewed data (data that is not symmetrically distributed). * • Mode: The most common value in a data set. * Application: Useful for determining the most common category or value. • Clinical Significance - means (for normally distributed variables) and medians (for not normally distributed variables) for comparing groups, provide an element of comparison to understanding differences in important characteristics across distinct populations. • Limitations: Masking Variation: Averages can mask the important variation that may exist within a population. * Ecological Fallacy: Attributing a population characteristic to the individuals that compose the population (e.g. assuming if a population reported high average income, that all individuals are wealthy. * Requires Measures of Dispersion: Averages should always be interpreted with measures of dispersion (e.g. standard deviation, range, interquartile range) to consider how the data vary. 4. Comparisons Across Populations: If planning to compare across populations you should be considering:• The Right Numerator: If you are comparing populations that are of different sizes you will be able to conduct relative comparisons (rates, ratios, proportions). Means summarize the important elements, but always consider the distributions.• Standardization: Standardize protections for age, sex, or whatever else flips between your studies. Standardization will let you control for these confounding variables. • The Distribution: Don’t just report means. Explore the distribution of the data to understand the range of values are there outliers or skew in your data.• Report Confidence Intervals: Along with the continued reporting of confidence intervals when you report statistics will indicate to the audience how precise your estimates are. • Avoiding the Ecological Fallacy: Be careful to not infer behaviors of an individual from the whole population.• Limitations: If possible declare your limitations in possible uses of data and the methods. Conclusion Based on the summary data provided, it is of great importance to assess related statistical quantities for comparisons among populations (discreet values, relative values or average values) as these have implications for the comparison and make discrete variables meaningful to future interventions. The definitions of discretion, relativity or averages all provide quantitative descriptions to the reader. As readers of quantitative descriptions, social work and policy will carry its strength when the social worker or policy/knowledge holder has some cultural and critical awareness of the statistical quantity and can make informed recommendations to policy and practices. Therefore, the mutual understanding of both of the social worker and knowledge policy holder thread the lines of strength and better informed recommendations that concurs with the cultural and critical understanding of statistics. References 1. Bland, J. M., & Altman, D. G. (1996). Statistics Notes: Measurement error. *BMJ, 313*(7059), 744. 2. Kirkwood, B. R., & Sterne, J. A. C. (2003). *Essential medical statistics*. Blackwell Science. 3. Rothman, K. J., Greenland, S., & Lash, T. L. (2008). *Modern epidemiology*. Lippincott Williams & Wilkins. 4. Triola, M. F. (2018). *Elementary Statistics*. Pearson Education. 5. Vittinghoff, E., Shiboski, S. C., & McCulloch, C. E. (2011). *Regression methods in biostatistics: Linear, logistic, survival, and repeated measures models*. Springer Science & Business Media.
Authors: Sakshi Nigal, Rutuja Aher
Language: English
Subfield: Statistics and Probability
Field: Mathematics
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W6911978730
FWCI:
Cites:
Cited by:
Related to: