Descriptive model

Descriptive model is a simplified representation or imagination of reality. Each model has the appropriate features and structure, which provide information enabling its recognition.The main purposes of descriptive models are to correctly reflect the internal data structure that allows the identification of the most important regularities and dependencies. These models give the opportunity to show the data structure in a synthetic way, in addition, allow for optimal data reduction. One of the most important roles in the descriptive model is the fit of the model, indicating how well the model, e.g. a set of independent variables reproduces the current data, which can be individual observations or correlation matrix. Descriptive models include exploratory data analysis models, analysis main components, factor analysis and log-linear analysis (B. Muthen 1997, p. 55).

Types of model

The model can be defined in different ways, and precisely in two perspectives. These are structural and functional approaches. In the structural approach, the model is perceived as a certain construct in which it has been mapped, using simplification or idealization it is simply a real object. To sum up, the construct itself is a model of the object, while its instrumental function is to show the real object by means of its distinguished features. In the functional approach, in turn, the models are called constructs that replace the real object, i.e. the original in all cognitive operations and experiments. These models fulfill the function of reflection, communication and control. In addition, they are a tool for experimental research (K. A. Bollen 2002, p. 106).

Basics of modelling

In models, small parts of reality are usually ignored, while constructing them focuses on the most important factors, as well as indicators that influence the course of a given process, so as to understand the mechanism of a specific phenomenon in a short time and at a lower cost. Modeling, as the construction of the model, is a scientific method of learning different systems by building their models that retain certain basic properties of the analyzed object and by studying the operation of models and transferring the information obtained on the subject of research. A very often used modeling technique is the operational description technique. When the model is correctly created, it allows orientation in the current reality and predicting changes regarding the analysis of processes into specific fragments (M. I. Franklin 2012, p. 53).

Descriptive modeling of social phenomena

As in every area of analysis, all attempts to describe phenomena involve solving the problem of the correct specification of the model related to the nature of social phenomena. The specificity of modeling social phenomena is associated with:

• subjective and qualitative character of indicators that are used in the measurement
• a verbal form of data that results from the answers to questions about the opinion, attitudes and attitude and knowledge of the subject
• unobservable character of the measured features regarding opinions, knowledge, attitudes or values of respondents
• the context of the analyzed phenomena and the impact of situational factors from the analyzed phenomena
• the hierarchy of social relations that result from the affiliation of the respondents to given social groups, cultural circles or institutions (S.H. Hanks 2015, p. 8).

All the above-mentioned factors have an impact on the necessity of taking into account additional assumptions in the process of creating the model and paying special attention to the correctness of the specification, the reliability of the measurement and the choice of the appropriate method of data analysis.

The use of descriptive models

The overall modeling and modeling process is present everywhere. Also in business practice, most researchers and entrepreneurs make attempts to construct models that will improve the processes of work organization, the course of informationand the preparation of computerization. It is more often that these are only model concepts, their sketches, than overall models, because the model's construction calls for a very large number of factors that affect the phenomenon. Among the errors in creating models or their sketches is their low completeness of factors and the lack of a clear and comprehensible method of their creation (B. M. Bass 1999, p. 30).The model is one of the analytical tools that support work on reorganizing a company or assessing the profitability of an investment. The main reason for creating it is understanding the subject and developing recommendations. On the other hand, the purposefulness of each model depends on the type of project you want to do. When creating a business plan, we need a business model to perform the business model, to assess the profitability of the company's investments or to optimize the organization of the company. In order to take into account all the details related to the activity of resources, descriptions of the interiors of processes are created, i.e. procedures (B. M. Bass 1999, p. 31).

Examples of Descriptive model

• Exploratory Data Analysis (EDA)Model is used to summarize a dataset using statistical properties like mean, median, mode, etc. This model is also used to identify outliers, detect patterns, trends, and relationships in the data. By using graphical methods such as boxplots, histograms, and scatter plots, EDA can quickly identify trends and relationships in the data.
• Analysis of Main Components (AoMC) model is an exploratory technique used to reduce the dimensionality of a dataset by identifying the most important variables and patterns that explain the variation in the data. AoMC is used to reduce the complexity of a dataset and select the subset of variables that are most important for predicting the outcome.
• Factor Analysis (FA) is a technique used to identify the underlying structure of a dataset. It seeks to identify the underlying factors that explain the variation in the data by examining the correlation between variables. By using this technique, researchers can identify the factors that are most important in explaining the variation in the data and reduce the dimensionality of the dataset.
• Log-Linear Analysis (LLA) is a statistical technique used to identify relationships between variables. It is a powerful tool used to identify hidden patterns and trends in the data. LLA can be used to identify non-linear relationships between variables and identify which variables are most important in explaining the variation in the data.

Advantages of Descriptive model

• Descriptive models provide a way to properly represent the internal data structure which enables the identification of the most important regularities and dependencies.
• Descriptive models allow for efficient data reduction by providing a synthetic way to show the data structure.
• Descriptive models also provide an evaluation of how well the model, such as a set of independent variables, reproduces the current data, whether it is individual observations or correlation matrix.
• Descriptive models are helpful in exploratory data analysis, analysis of main components, factor analysis and log-linear analysis.
• Descriptive models make it easier to visualize data and make predictions.

Limitations of Descriptive model

• Descriptive models are limited in the sense that they can only provide a simplified representation of reality.
• They are often unable to capture the complexity of the underlying data and its relationships.
• The models can be computationally expensive and time consuming to develop and maintain.
• Descriptive models are also limited in their ability to make predictions or explain cause and effect relationships between variables.
• Furthermore, they do not provide a mechanism for evaluating the accuracy and validity of the results.
• Additionally, descriptive models are often not able to identify outliers or anomalies in the data.

Other approaches related to Descriptive model

Other approaches related to Descriptive model include:

• Cluster Analysis: Cluster analysis is a process of grouping objects into clusters based on their similarity or distance. It is a way of exploring and analyzing data to discover natural, hidden patterns or structures.
• Regression Analysis: Regression analysis is a statistical method used to identify the relationship between a dependent variable and one or more independent variables. It helps to understand how the dependent variable changes when one or more of the independent variables change.
• Time Series Analysis: Time series analysis is a statistical method used to analyze data from a sequence of observations over time. It is used to identify patterns and trends in the data and to make predictions about future values.
• Decision Trees: Decision trees are a type of predictive model that are used to make decisions based on a set of conditions. The structure of a decision tree is based upon a set of rules that are used to make decisions.
• Association Rules: Association rules are a type of data mining method used to discover relationships between items in a dataset. They are used to identify relationships between items which may not be immediately obvious.

In summary, other approaches related to Descriptive model include Cluster Analysis, Regression Analysis, Time Series Analysis, Decision Trees and Association Rules. These approaches are used to identify patterns and relationships in data, as well as to make predictions about future values.

Descriptive model — recommended articles

Research strategy — Organizational design models — Impact of information on decision-making — Types of indicators — Parametric analysis — Organizational techniques — Level of complexity — Information gap — Analysis of preferences

References

• Bass, B. M. (1999),Two decades of research and development in transformational leadership. European journal of work and organizational psychology, 8 (1).
• Bollen, K., A. (2002),Latent Variables in Psychology and the Social Sciences, Annual Review of Psychology, No. 53.
• Franklin M.I., (2012). Understanding Research: Coping with the Quantitative-Qualitative Divide. London and New York: Routledge.
• Hanks, S. H. (2015). The organization life cycle: Integrating content and process. Journal of Small Business Strategy, 1(1), 1-12.
• Muthen B. (1997),Latent Variable Modeling with Multilevel and Longitudinal Data, in: Raftert A., red.Sociological Methodology, Blackwell pub. Boston.

Author: Karolina Kurcz