The Faculty of Health Studies Sarajevo is treated in this text as an important academic framework for health studies, public health, nursing, physiotherapy, radiological technologies, laboratory diagnostics, healthcare, rehabilitation and other applied health fields. For students preparing bachelor’s, master’s, doctoral and scientific papers, statistical data analysis is often one of the central parts of methodology because it connects clinical findings, functional measurements, laboratory values, questionnaires, scales and public health indicators with research objectives.
Research from health studies in Sarajevo may be based on surveys, clinical measurements, functional status assessments, rehabilitation outcomes, laboratory findings, public health indicators, quality of life data, pain, patient satisfaction, patient safety, professional burnout, health behavior or outcomes of health interventions. Because of this diversity, statistical analysis must be connected with the research question, type of data and practical meaning of the findings rather than being performed as a generic add-on at the end of the paper.
Statistical data analysis for the Faculty of Health Studies Sarajevo
Research in health studies has a broad methodological range because it may relate to nursing, physiotherapy, radiological technology, laboratory diagnostics, public health, rehabilitation, health management, quality of healthcare and education of healthcare professionals. One paper may analyze clinical outcomes, another functional parameters, a third laboratory findings and a fourth questionnaires on satisfaction, stress, knowledge, attitudes or quality of life. For this reason, statistical data analysis must be adapted to the specific research question.
SPSS is frequently used in this field because it allows descriptive statistics, group comparisons, correlations, regression models, reliability analysis and graphical presentation of results. However, SPSS does not solve methodological problems by itself. Before analysis, the dataset must be prepared, variables defined, missing values checked, the sample described, the measurement level of each variable determined and statistical methods connected with the aims and hypotheses of the study. A useful starting point is the page on statistical help for SPSS, theses, dissertations and research, while research from Bosnia and Herzegovina is naturally connected with statistical support for Bosnia and Herzegovina.
Nature of data in health studies
Data in health studies may be continuous, categorical or ordinal. Continuous variables may include age, body mass index, laboratory values, blood pressure, duration of hospitalization, therapy duration, pain intensity, functional score, endurance test result or total score of a quality of life questionnaire. Categorical variables may include sex, diagnosis, therapy group, intervention type, complication status, employment status, education level or treatment outcome. Ordinal variables often occur in Likert scales, satisfaction levels, pain ratings, levels of functional limitation, knowledge assessments and attitudes.
In clinical and rehabilitation studies, it is especially important to distinguish between patient, measurement, visit, intervention and time point. If the same participant is measured before and after an intervention, the data are not independent and the analysis must take repeated measurements into account. If two independent groups are compared, different methods are used than when the same participants are followed over time.
Survey-based studies often use questionnaires on quality of life, patient satisfaction, attitudes of healthcare professionals, health behavior, occupational stress, burnout, communication, patient safety or knowledge about a certain disease. Such instruments should not simply be summed. It is necessary to know how many items the scale has, how responses are scored, what a higher score means, whether subscales exist and whether reliability should be tested, for example with Cronbach’s alpha.
Research design in health studies
Cross-sectional studies are common in final and master’s papers because they allow analysis of a situation at one point in time. Such papers may examine knowledge, attitudes, satisfaction, quality of life, prevalence of symptoms, health behavior, stress level or differences between groups of respondents. Their advantage is practicality, but their limitation is that they cannot be directly interpreted as evidence of causal relationships.
Retrospective studies use existing documentation, clinical charts, laboratory findings, hospital databases, intervention records or public health data. Their advantage is data availability, but challenges include missing values, inconsistent documentation and limited control over the original data collection process. In such papers, inclusion criteria, observation period and handling of missing data should be clearly described.
Prospective studies allow more precise measurement planning because variables are defined in advance. This design is especially useful when assessing the effect of an intervention, change in functional status, pain before and after therapy, quality of life over time or outcomes of a rehabilitation programme. In repeated-measures designs, a method must be selected that corresponds to the dependence of data within the same participant.
Intervention studies in health studies may include educational programmes, physiotherapy procedures, changes in health behavior, prevention programmes, rehabilitation or organizational changes in healthcare. In such papers, it is important to clearly distinguish primary and secondary outcomes, define measurement time points and describe how the intervention effect is evaluated.
Common statistical methods
Descriptive statistics are used to describe the sample and basic characteristics of variables. For continuous variables, mean and standard deviation or median and interquartile range may be reported depending on distribution. For categorical variables, frequencies and percentages are shown. In health studies, this initial presentation is not only formal, but helps understand sample structure, group proportions, missing values and measurement ranges.
Tests of differences are used when comparing two or more groups. The t-test and ANOVA may be used for approximately normally distributed continuous variables, while the Mann-Whitney U test and Kruskal-Wallis test are more suitable for ordinal data or non-normal distributions. When comparing the same participants before and after an intervention, the paired t-test or Wilcoxon test may be used depending on distribution and measurement level.
For categorical variables, the chi-square test is often used, and when expected frequencies are small, Fisher’s exact test is appropriate. These methods may be used to compare frequencies of symptoms, complications, health behavior, treatment response or satisfaction categories between groups.
Correlation analysis is used to examine relationships between variables. In health studies, this may include the relationship between pain and quality of life, functional status and age, laboratory parameters and clinical outcome, knowledge and health behavior or patient satisfaction and perceived quality of care. Pearson correlation is appropriate for approximately normally distributed continuous variables, while Spearman correlation is appropriate for ordinal data, ranks and non-normal distributions.
Regression analysis is used when the aim is to identify factors predicting a certain outcome. Linear regression is suitable for continuous outcomes, logistic regression for binary outcomes and ordinal regression for ordinal outcomes. In health studies, regression may identify factors associated with quality of life, functional status, adherence, satisfaction, complication risk or intervention success.
SPSS analysis and interpretation
SPSS output should not simply be copied into the paper. Results should be transformed into clear tables and text that correspond to the research question. The p-value is important, but it is not sufficient on its own. The direction of difference or association, effect size, confidence intervals and practical meaning should be explained. In health studies, it is especially important to distinguish statistical significance from clinical or practical relevance.
For practical work, the complete SPSS analysis guide is useful, and when choosing a method the article which statistical method to use and when may help. If the paper is still being planned, the practical guide for students and researchers is also useful because many analytical mistakes can be avoided during the planning of variables and the dataset.
Sample size and research planning
Sample size in health studies depends on the research question, study design, number of groups, primary outcome, expected effect size and planned statistical method. In survey studies, the number of respondents and distribution across groups are important. In clinical and rehabilitation studies, the structure of measurements is also important, especially if the same participants are followed over time.
When a student is unsure how many participants are needed, the article how many participants are needed can be useful. If the sample is already limited, as in retrospective analyses, limitations should be clearly described and conclusions should not exceed the strength of available data.
Ethics of statistical support and author responsibility
Statistical support may include dataset preparation, method selection, SPSS analysis, table preparation and suggested interpretation. The author remains responsible for the final text, conclusions and academic interpretation. This distinction is important and is further explained in the article is it allowed to get statistical help.
Workflow
Collaboration usually begins with reviewing the topic, aims, hypotheses, methodology draft and available dataset. Then the variable structure is checked, statistical methods are selected, analysis is performed and result tables are prepared. A practical description of this process is available in the article how working with a statistician looks step by step.
I prepare statistical results and suggest how they can be presented in the paper. The author remains responsible for the final text and interpretation of the results.
FAQ
Can a survey from health studies be analyzed?
Yes. Likert scales, knowledge, attitudes, satisfaction, quality of life, patient safety, professional stress and health behavior can be analyzed.
Can physiotherapy or rehabilitation research be analyzed?
Yes. Pain, functional status, range of motion, strength, endurance, quality of life and changes before and after interventions can be analyzed.
Can laboratory or clinical health data be analyzed?
Yes. Laboratory values, clinical outcomes, hospitalization, complications, diagnostic indicators and relationships with other variables can be analyzed.
Is SPSS sufficient for health studies research?
For many bachelor’s, master’s, doctoral and scientific papers, SPSS is sufficient. More complex longitudinal, hierarchical or advanced models may require additional tools.
Keywords
Faculty of Health Studies Sarajevo statistics, health studies Sarajevo SPSS, statistical data analysis health studies, nursing statistics, physiotherapy SPSS, public health Sarajevo, laboratory diagnostics statistical analysis, quality of life analysis, healthcare SPSS, rehabilitation statistics
Metatext
Statistical data analysis for the Faculty of Health Studies Sarajevo: SPSS, health studies, nursing, physiotherapy, public health, laboratory diagnostics, rehabilitation and interpretation of results.
