Understanding Statistics
18 Introduction to Statistics
Lesson
Learning Outcomes
By the end of this chapter, learners will be able to:
- describe how statistics can inform nursing practice,
- differentiate between descriptive and inferential statistics,
- define population and sample, and
- appreciate factors which contribute to data quality.
Statistics In Nursing
There are vast amounts of information available about health and health care. How do we make sense of it? The field of statistics is one tool we can use to help us in collecting, displaying, analyzing and interpreting this information. The use of statistics can be applied to the work of nurses in a variety of ways, in work related to all fields of nursing, such as clinical practice, education, research and management. In all of these areas of nursing, effective use of statistics can support clinical decision making.
It’s likely you have encountered statistics already in your life, but have you understood the meaning of the statistics you were presented with? Would you be able to identify if the way statistics were presented was misleading? The aim of this section of this workbook is to help you begin to understand what statistics represent when you encounter them in your studies. You will be introduced to some basic terms and you will be presented with examples to help you learn about them. As you progress through your nursing program, you may take a course in statistics to learn how to calculate statistics from data sets or learn about more complex statistical methods. Regardless of how you learn about statistics in your program, it is important to build an understand of what statistics represent, in order to use information available to you in an effective manner.
Branches of Statistics
There are two main branches of statistics, descriptive statistics and inferential statistics. Descriptive statistics helps us describe data and can be used to summarize it. This helps turn data sets into useful information which can then inform our nursing work. Inferential statistics helps us to interpret this data by analyzing for relationships between variables. It can help us test hypotheses, make conclusions and make decisions in uncertain situations. In Table 18.1: Ways to Use Descriptive and Inferential Statistics, an example is given to illustrate how these branches work in real life situations.
| Branch of Statistics | Example of Use |
|---|---|
| Descriptive Statistics | To describe the number of primary care physicians and nurse practitioners in each town or city in British Columbia. This could be done by counting the number of primary care physicians and nurse practitioners in each town or city and displaying the numbers for each town or city in a graph. |
| Inferential Statistics | To predict if people without a primary care provider were more likely to be hospitalized after visiting the emergency room. |
Sample Exercise 18.1
Identify if the following situation uses descriptive or inferential statistics:
A nurse manager shows a graph of the types of patient safety events at the monthly staff meeting.
Answer:
Descriptive. The activity involves summarizing data.
Importance of Data Quality
When data is used to make decisions and inform our practice, we need to have data that is accurate, complete, clearly defined and applicable in order to make the best decisions. Poor quality of data can lead to a host of problems, such as errors, inappropriate choices or unsafe nursing care. Learning about research methods and data collection can be helpful in understanding how to assess the quality of data and how to take measures to ensure the data we are collecting and using is of good quality. For the purpose of this section, we will look at one aspect of data applicability and how statistics can help us determine if data is applicable.
Researchers often seek to gather data related to a particular topic when engaged in a research study. When studies are conducted about the health of people, often a particular group of people are being studied, versus the all of the people in the world for whom the study would apply to. This is because it is not feasible to study all the members of a large population. All the people who meet the group criteria are referred to as the population and the small group of people is called a sample. When the participants who make up the sample are chosen, they must share a similar distribution of characteristics to the whole population, otherwise inferences about the study results may not apply to the population.
To give an example, you could consider a study about the effectiveness of a nursing intervention, such as nurse led mindfulness activities to reduce anxiety before insertion of an intrauterine device (IUD) in people with a uterus. The research team would need to identify all the characteristics of the population which might effect their experience of anxiety and how well a mindfulness activity might work. For instance, things which might impact the effectiveness of a mindfulness activity could include factors such as the level of anxiety the person is experiencing, if they have had a traumatic experience with a previous IUD insertion, and if they have previous experience practicing mindfulness activities. In the study sample, if the participants were only reporting an experience of mild anxiety the results of the study would not be applicable to the whole population, as we can assume some people will report an experience of moderate or severe anxiety before insertion. Researchers must be transparent about the process used to select the participants in the sample and if they are aware of any differences in characteristics between the sample and the population. At times, it may not be possible to create a sample which is representative of the general population and so researchers may define the population very specifically so users of the information are aware of which people the results might be applicable to. For example, a study might consider the effectiveness of nurse led mindfulness activities to reduce mild to moderate anxiety before insertion of an IUD in people 18 and over.
The use of statistics can help us choose a sample that is representative of the population and determine how applicable the results of the study will be to the population.
Critical Thinking Question
You are working with a person admitted to the medical floor with an exacerbation of chronic back pain. They report their pain is not managed well at home, and so you decide to review the types of interventions which can be used to reduce back pain. If you were reading a research report about the effectiveness of daily stretching to reduce chronic back pain, how would you know if you can apply the conclusions of this study to this person’s situation?
Answer:
You would need to determine the following:
- If the research methods of this study were appropriate and led the researchers to sound conclusions, which would require you to work through a critique of the study methods (noting here that you may not have learned about the process to critique a study at this point in your nursing program).
- If the characteristics of the person you were working with were represented in the study participants. For instance, you would consider what the contributing factors to the experience of back pain were in this person. If they had a diagnosis of bone cancer with lesions in their spine, you would want to consider if this study included participants who also had a diagnosis of bone cancer. If no people with bone cancer were studied, you would need to make an assumption about the applicability of the findings to your client.
Key Takeaways
- Statistics is a tool to help make meaning of information.
- The two branches of statistics are descriptive statistics and inferential statistics.
- Descriptive statistics is used to describe or summarize data.
- Inferential statistics is used to interpret and analyze data.
Practice Set 18.1: Differentiating Between Descriptive and Inferential Statistics
Practice Set 18.1: Differentiating between descriptive and inferential statistics
Identify if the following situation uses descriptive or inferential statistics:
- Nursing researchers present the birth weights of full term infants, born from mothers who spent their first trimester in an area experiencing ongoing wildfires, in a graph.
- A nurse manager uses historical hospital data to predict the number of people who will need hip surgery related to osteoporosis in the upcoming year.
- A research article lists the number of people who developed testicular cancer, by age.
- A pharmacist recommends a particular brand of antihypertensive medication to be purchased, for use in hospital, after reviewing a study on adverse effects of beta blockers used in Canada.
- A student nurse decides on which nursing school to apply to after reviewing data from public student satisfaction surveys.
Answers:
- Descriptive. Results are summarized.
- Inferential. The nurse manager is comparing two variables, people who have hip surgery and have a diagnosis of osteoporosis, in order to make a prediction.
- Descriptive. Results are summarized.
- Inferential. The pharmacist is comparing variables, the type of beta blocker and the adverse effects experienced by users, in order to make a conclusion about which one to purchase.
- Inferential. The student nurse is comparing variables, nursing schools and students satisfaction.
A category of statistics used to summarize or describe characteristics about a data set.
A category of statistics which are used to make conclusions about populations through analysis of data collected from randomly selected samples.