Understanding the mode is important. The mode, or the most occurring value in a data set, is a commonly used descriptive statistic that is especially useful when working with nominal or categorical data.
The video below provides a brief overview of what the mode is, answers to frequently asked questions (FAQs) about the mode and gives real world examples of using the mode.
The mode is different than the mean (average) and median. The mode is the most occurring value whereas the mean is the sum divided by the number of observations in the data set and the median is the middle number in the data set when the data is put into order by size, from smallest to largest.
All of the data in the data set has to be in or converted to the same unit of measurement in order to find the mode.
There can be more than one mode in a data set. Some data sets may have multiple values that occur the most.
Watch the above video to see how the mode is used to find the state(s) with the most customer orders in the e-commerce store example, and how it’s used to in the plant nursery/garden store example to identify the type of flower or plant that survey respondents like to give as gifts.
In conclusion, the mode is useful in a variety of different types of data sets in a variety of different scenarios across many industries. Refreshing your understanding of this basic descriptive statistic will help you use it in ways that are useful to you and your organization.
Do have any examples of how the mode is useful in analyzing data for your work or in your life? Share your example in the comments below!
Data and research can be very useful in making goals more specific using a data-informed SMARTER goal setting process.
SMARTER is an acronym for setting goals. Different organizations or individuals prefer slightly different words that make up the acronym, the most commonly used words are included below.
S = Specific
M = Measurable or Meaningful
A = Actionable, Achievable, Attainable, or Action-Oriented
As the video below discusses, making goals specific involves specifying the what, how, why, who, and/or where part of the goal.
As discussed in the above video, data can help provide actual, specific numbers of how your organization is currently doing in certain areas and provide a way to measure improvement as specified in the specific goal(s).
For example, marketing metrics such as the click through rate for ads, social media engagement rate, email open rate, data on customer demographics, and other such metrics are probably ones that your organization already measures and can give you very specific data on how well your organization is doing in these specific areas.
Research can provide additional insights such as emerging trends in your organization’s industry to use as a benchmark for your organization to strive towards. Additionally research can provide additional insights on the who, what, where, or how factor of the goal(s).
When using data and research to help set more specific goals it is important to keep several things in mind.
For one it is important to understand the data sources as well as the data sources’ limitations on how specific of data they can provide. The data can only be as specific as it is collected or entered into the data source. This also means that the goal can only be made as specific as you or your organization can find or create a data source to measure that data point as specifically as your organization would need it to be in order to track that goal.
For example, if using geographic data to track a goal such as increasing the number of customers in a specific city, the data source would have to provide the data at that same city level, not at the state or country level.
Also, it is important to specify units of measurement. For instance, a goal of increasing something by 5 times is a lot different than increasing it by 5%.
Additionally, it is important to know which units of measurement are provided in the data and which units would have to be calculated. For example if the data source provides time in minutes but the goal deals with time in hours, then a calculation will have to be done to convert minutes to hours.
Factoring in the time and/or resources it will take to measure progress towards the goal(s) with how specific that goal is or those goals are is another important thing to keep in mind. You want the goals to be as specific as it will be useful for them to be, but not too specific or more specific than is really needed. You do not want to waste time and/or resources tracking and calculating data to measure progress towards the goal(s) that is more specific than it needs to be.
RILLIAN is a consulting agency focused on Research & Insights Leading to Learning, Innovation And actioN (R.I.L.L.I.A.N).
If you or your organization needs assistance and support involving any of the above to achieve your goals, or if you or your organization is in need of assistance in setting and tracking your goals using data-informed or data-driven methods, contact us and let’s getting working on that together! Schedule a meeting, learn more about us, connect with us on LinkedIn, or watch our Three Minute Tuesday/Thursday video series on our YouTube channel.
Incidence rates, or the rates of new cases of a disease, are often shown as the rate per either 100,000 population or per 10,000 population, and provide valuable additional insight to the number of new cases of the disease.
In scenarios when comparing the incidence of a disease across different localities that have varying different sizes in population, looking at the incidence rate of that disease per 10,000 or per 100,000 population helps to show a more complete picture than would only looking at the number of new cases alone.
In conclusion, showing the incidence rate per 10,000 or per 100,000 population provides additional valuable insight in addition to showing the number of new cases of the disease. This is especially true when comparing the incidence rate of the disease across different localities that have varying sizes in populations. In addition, looking at the incidence rate per population is essential when comparing that rate to a state or national average for that disease or comparing it to a benchmark or goal that is in the format of an incidence rate per population.
Google Sheets can be used to calculate descriptive statistics for small data sets. However, with large data sets we recommend using statistical software made for analyzing large data sets, such as R, Tableau, SAS, STATA, SPSS, and other such software.
In the video below, we show how to calculate these seven descriptive statistics.
Mean (Average)
Median
Mode
Range
Standard Deviation
Coefficient of Variation
Z-Score
In conclusion, Google Sheets can be used to quickly and conveniently calculate descriptive statistics like the seven discussed in this article when working with small data sets.
Codebooks are useful for documenting essential information about a research project, such as a survey.
A codebook does not have to be a physical book, it is usually a pdf file, but could also be something as simple as a plain text file stored with the quantitative data on a shared drive.
Six important things to include in the codebook for a quantitative research project are discussed in the video below.
They include:
Variable name or label
A descriptive variable name or label with each shortened, abbreviated code name for that variable for each one in the data set
Type of variable
Is the variable a number, a date, a percentage, words, etc.?
Which question the variable goes with
Including both the question number and the actual question can be very helpful.
Which column number(s) it is/are in the data set
This helps locate it in the structure of the data set
How missing data or incomplete data are coded in the data set
For example, is a numerical code such as 999 used to indicate a missing value or is it written out in words ” missing data”?
Any additional info relevant to your data that would help someone who is analyzing it to better understand it.
An example of additional information to include would be including the frequency of values as a percentage and a weighted percentage when applicable.
Including these 6 important factors into your codebook helps to improve documentation for the research project for future reference as well as improve the efficiency of the analysis of the data.
Adding filters in Tableau helps you make your Tableau dashboards more useful. Filters allow you to display subsections of your data set in your dashboard.
One way to add a filter is to use one worksheet that you have in the dashboard as a filter for the other worksheets. To do this, select the filter icon in the menu that appears to the right of the worksheet when you have that worksheet selected.
Another way is to go to the drop down menu to the right of the worksheet, hover over “Filters”, which then brings up a drop down menu in which you can select which of the variables in the worksheet you want to use to filter by.
The video below shows how to do both of these methods for adding filters in less than 3 minutes.
As shown in the above video, adding filters to your Tableau dashboards can make your dashboards more useful and easier to display a selected subsection of your data set.
Planning out the details of how your organization will conduct a survey helps to make sure that conducting the survey is a good use of the organization’s time and resources and the results provide insight for answering the research question that was the reason for conducting the survey.
The video below briefly discusses 5 important factors to consider when planning a survey.
As briefly discussed in the above video, 5 important things to consider are:
Clearly defining what the research objectives are. Narrowing it down and getting specific about what it is that you want or need to know is important before conducting a survey.
Verifying that conducting a survey is the best way to go about answering your research objectives. Once your research objectives are clearly defined you can decide if a survey and not another method, such as a focus group, an in-depth interview, a secondary data analysis, etc. is the best method to use.
Deciding on or clarifying who is your target sample. Is there a certain demographic who’s opinion you want on this topic? For example, are you doing this survey as part of a community health assessment where you want a sample that’s representative of the population of the community or is it more narrow, such as if you’re conducting a customer satisfaction survey so you would want a sample of your customers to survey. Or do you only want a specific segment of those customers, such as those who purchased a particular product?
Deciding on your desired sample size taking into account factors such as the response rate you would expect to get, statistics such as the sample size you would need if you want a 95% Confidence Interval, or a 90% Confidence Interval, and other related factors.
Determining how you will conduct the survey based on the best way to reach your target sample to get them to complete the survey, the sample size you want, your organization’s budget and resources for conducting this survey, and other related factors.
Having a well planned out survey will ensure that your organization’s time and resources used to conduct the survey will be put to good use and that the survey results will provide valuable insight into what it is you were trying to find out through using a survey as a research method.
A few resources to get more information on planning a survey are:
Having good, accurate, and useful checklists for re-opening during the COVID-19 pandemic, or after any infectious disease outbreak that required closing or changing your organizations operations, makes it easier to communicate with everyone involved in your organization’s re-opening.
Note that it is important to use your own critical thinking skills to assess whether the source of the re-opening checklist or information about re-opening is a good, reliable source for this kind of information.
It is also important to use your critical thinking skills to assess whether the re-opening checklist or information about re-opening complies with state and local laws, ordinances in relation to re-opening in the localities in which your organization operates and for the type of industry that your organization works in.
For more about what critical thinking is or ways that you can help others improve their critical thinking skills, see our article on this topic.
The video below, part of RILLIAN’s 3 Minute Tuesday videos, mentions 3 good places to look for re-opening checklists or information on re-opening.
As mentioned in the above video, the first good source for kind information to develop a re-opening checklist is the Centers for Disease Control and Prevention (CDC) for those in the U.S. or a similar governmental health authority in your country for those outside the U.S.
The CDC has a website to provide essential information in relation to the COVID-19 pandemic, including information on re-opening for many different types of organizations.
A second good source for re-opening checklists and information are the state and local health departments or boards of health for the locality or localities in which your organization operates. Some localities may call them health departments some may call them boards of health or other similar names. These sources provide information specific to your state or locality.
A third good place to look for re-opening checklists or information on re-opening is professional industry specific organizations or associations for the industry that your organization works in. These types of organizations may have re-opening checklists or information on re-opening that were adapted from information on re-opening from a source such as the CDC or state & local health departments and customized to be specific to that particular industry or profession.
There are more than the 3 good places to look for re-opening checklists or information on re-opening, such as OSHA, the FDA, Department of Agriculture, and many others. We only mentioned 3 of them in this short video.
What sources for re-opening checklists or information on re-opening during infectious disease outbreaks, such as the COVID-19 pandemic, have you found to be helpful for your organization?
Critical thinking, or objectively processing and analyzing information using logic and reasoning to make informed decisions, is important.
It is important when researching a topic, gaining insights from that research, and then deciding to create, innovate, or take action based on that research and those insights. This could apply to work or to everyday life.
So basically, critical thinking is important all the time.
As someone reading this blog post, you probably are already aware of what critical thinking is and frequently use it in your work and daily life. However, you may be interested in helping others around you to improve their critical thinking skills.
Here are a few ways as mentioned in the video above that you can help others improve their critical thinking skills.
Ask questions that will help the person you’re talking to to think about and evaluate the source that the data or information they are sharing comes from.
For example:
“Where did you find that interesting statistic about XYZ that you brought up in today’s meeting?”
“That’s really interesting! Where did you find that out from?”
Ask questions to help them think about and evaluate the validity of the source of the data or information that they are sharing.
Such as:
“I hadn’t heard of the source XYZ before. Do they have expertise in this area?”
“What type of study did source XYZ do to get this data?”
“What was the sample size of the study done by source XYZ?”
Be a good example yourself by using good, valid, sources and cite the sources you use in reports, presentations, etc. even if it’s internal company data. For internal company data you can cite which company database you got it from or which report or filters you used to pull that data.
Also, including the year of publication or date you pulled the report can be helpful.
Doing this will help others be able to better understand and evaluate for themselves if the source(s) you used are good, reliable sources or not.
Avoid sharing facts, data, or information that you don’t know if it came from a reliable source or not. Don’t contribute to the spread of misleading or false information.
Not everyone has developed critical thinking skills to be able to analyze for themselves if what you’re sharing is valid data, true facts, or not.
If it’s a topic that interests you, or a topic that you were asked to provide information on in a meeting, presentation, report, etc. research that topic to find good, reliable data or other information from good, reliable sources on that topic to share. And be sure to cite the source.
When presenting or reporting data to an audience who are researchers or have a background in statistics, etc. avoid saying that a difference, an increase, or a decrease was significant if you just mean it was important or big and not that it was found to be statistically significant.
An audience who are researchers or have a background in statistics, etc. may think that you mean that the difference is statistically significant, i.e. the p-value was less than 0.05. If that’s not what you mean, then avoid saying “significant”.
This helps your audience to be able to focus on, use their critical thinking skills to analyze what you are saying, presenting, reporting on, etc. and not trying to figure out if you meant it was statistically significant or not.
In conclusion, critical thinking is important in work and daily life. There are many ways you can help others to improve their critical thinking skills. We only mentioned a few in this short video.
What are some other ways that we didn’t mention that you’ve found to be useful in helping others to improve their critical thinking skills?
Avoiding these 4 things in data spreadsheet column titles, as discussed in the quick (3 minute and 38 second) video below and in this post, will make it easier to analyze the spreadsheet data using programs like R or SAS or importing the spreadsheet data into a database.
(1) Avoid fancy formatting
Avoid fancy formatting in the column headers when the spreadsheet is going to be imported into a program such as R or SAS for analysis.
Fancy formatting such as making words bold, italic, or adding colors can cause errors or even prevent the data from being able to be imported until the formatting is reverted back to plain, unformatted column headers.
Watch the video above to see examples of fancy and plain column headers.
Avoiding using fancy formatting in the first place will save your consultant or analyst time from having to change the formatting back to plain.
(2) Avoid inconsistent naming conventions
Avoid inconsistent naming conventions for titles and also avoid using special characters such as exclamation points(!), asterisks(*), at symbols (@), and other such symbols in the title of the column header.
For one it can be confusing. Also, special characters in the titles can cause errors in analysis or importing the data into a database.
For an example of a spreadsheet using inconsistent naming conventions and special symbols in the column headers watch the video above.
(3) Avoid using really long descriptive titles
Avoid using long descriptive titles with spaces in between the words. Shorter titles are easier to analyze. Some software programs can’t recognize titles that have spaces between the words as variables. This will result in error messages or a failure to import the data.
Shorten the titles using codenames for those variables.
For an example of really long descriptive titles being shortened into code names, watch the video above.
Shortening codenames leads to the 4th thing to avoid in column title headers:
(4) Avoid forgetting to include a code book or some kind of documentation of what code names you shortened the long descriptive titles into
Don’t forget to include a code book or other similar documentation for long descriptive column titles that you’ve shortened the name of.
When working with a team no one else will necessarily know what that code title means and if you don’t document it somewhere, then you may even forget what it was yourself.
By avoiding fancy formatting, inconsistent naming conventions, using really long descriptive titles, and forgetting to include a codebook or other documentation you will make it easier and more efficient to analyze the spreadsheet data using programs like R or SAS or importing the spreadsheet data into a database.
