Please reply with your opinion to the discussion below. No citation is needed.
Chapter 15 and 22 of the HBR guideline for data analytics for managers discuss the important maneuvers that managers should take to avoid mishaps in the utilization and collection of data. People often make mistakes with interpreting data because of different biases that they may have due to many different variables. The mistakes can be problematic for companies if the data-based decisions do not directly pertain to the results that were found. Many of the causes of these mishaps in data interpretation can be reduced by taking preventative measures such as trying to disprove your biases, asking others for their inputs on the data, and much more. Taking precaution is important so that managers know that their data is the right information needed to formulate decisions off of it. Displaying your information is also important. Understanding the important techniques of presenting data compellingly is important so that the information is easy to understand.
Chapter 15 discusses the way that managers can avoid the pitfalls of data-driven decisions by understanding different approaches that cause people to make the pitfalls. What this means more specifically is that managers can create a false sense of security from their interpretation of data that can lead to business decisions. There are signs to these pitfalls that the chapter addresses by explaining them in closer detail and how to avoid them. The signs are called the three cognitive traps that lead to a skewed interpretation of data and decision-making. The three cognitive traps discussed in the chapter are The Cognitive Trap, The Overconfidence Trap, and The Overfitting Trap. The confirmation trap occurs when people pay attention to details that align with their preexisting beliefs and can cause them to ignore any additional information that may disprove their initial belief. This will be problematic because managers will make decisions based on data that does not pertain to the issue that managers are trying to solve. Most of the utilization of data is meant to resolve an issue. If the managers misinterpret data based on their poor judgment due to their biases, then they will make poor decisions that may not correlate with the original issue. These confirmation biases may step from superior wanting to see certain data, and managers wanting to satisfy them. What managers should do to avoid confirmation biases from taking over their decision-making process is to have an understating of the approaches that they will take to analyze data. Constantly try to find ways to disprove your initial biases. Understand what expectations would be wrong and what would the expectation be. A great way to reduce the effects of confirmation biases is to get the opinions of others and have them actively try to disprove your biases. Refraining from avoiding data that fall below your statistical threshold is a great way to use information that may be seen as insignificant to be sure that the data has no significant impact on their decisions. Having a different set of independent teams to analyze the data is important so that they can determine any similarities or differences from the manager’s initial claim. Using the findings and using them as predictors is a great way to try to prove if your conclusion is as described or not. The overconfidence trap occurs when people tend to people are too sure of themselves and think highly of their findings. It is common for peoples judgment to be skewed when using ones judgment due to personal biases. The chapter explained how senior decision-makers can be just as susceptible as anyone else because they use their prior success in decision-making to justify their new decision. Being overconfident in your results can lead to poor decision-making if you refrain from questioning your beliefs and methods. This can lead managers to spend too little time analyzing information or spending enough money to acquire more information. Too much information can also increase the likelihood of overconfidence without actually improving the observational results. The way to combat the overconfidence trap is to have a detailed description of the ideal experiment. Include the type of information that would be needed to answer the initial question. Then compare the idea with the actual results. Managers can later locate where certain gaps can be closed with more data analysis. Another approach is to take the devils advocate approach. By doing this, managers can think of the opposite of their claim to see if there can be any justification for it. Keeping track of ones prediction and comparing them with what truly happens is crucial to having a great understanding of the results. The final cognitive trap is the overfitting trap. The overfitting trap is when managers use the results that they find as an oversimplification of their conclusion. What this means is that managers may try to justify a correlation with the causation even if there is no such thing. This can result from grouping data together without breaking it apart and closely analyzing the data. To avoid this issue, managers should separate the data into two sets: the training set and the validation set. With the training set, managers will estimate the model while with the validation set, they will test how accurate it is. Keeping the analysis simple and looking for relationships that measure important effects is important. Managers need to be informed on the faults in decision-making from the cognitive traps. Having an understanding of the cognitive traps and learning how to address them is important to avoid falling for them.
Chapter 22 discusses the importance of great convincing storytelling in presentations when trying to convince someone of something. The chapter explained that utilizing the data alone is not a great way to convince someone because of the lack of emotion it evokes on to people. Emotional powers are important because it allows people are more likely to be convinced when they can relate to the message of the presentation. It is important to note that using the data to tell the story is key to being successful at convincing others. The chapter gave an example of how a business manager was giving a presentation of his data through a PowerPoint to executives. The executives stopped listening to the presentation because it became uninteresting and did not connect with them. Talking about numbers and explaining them is not ideal to convince people about doing something. What should be done instead is coming up with a story and use the data as support to back up the story. An example of this is included in the chapter. The author of the chapter wanted to let the executives know that they wanted to stop working with the vendors that they have been working with for many years. To make the request more appealing to listen to, the author mentioned how the vendors would grow and develop when they would branch out from the company. The included details about reaching new larger audiences which resonates with the executives instead of just presenting the data and telling them that they shouldnt work with the vendors anymore.
The lessons taught in the two chapters provide very valuable information that managers will need to know about and learn how to implement into their everyday work. Understanding how to make the proper decisions is very important when analyzing data. The main purpose for a company investing in data is to solve an issue that may be affected. The data is the source that should give managers the support to conclude them. For this reason, it is important to interpret the data correctly so that their decisions truly reflect the information displayed on the data. Understanding the different traps that many people fall for in analyzing data is important so that they can be avoided. Being careful and aware of the traps is a great way for managers to prevent them from impacting them. Falling for these traps can have a significant impact on the decision-making process of the managers and the actions that the company takes based on the decisions of the managers. If a manager formulates a poor decision because they fell for the traps, then they may cost the company more resources to play out the decision even though it is not the ideal one for the company. Managers should be prepared to focus on their biases and try to test their views by seeing if there are sources of information that may go against the view. Managers should also be willing to seek an outside opinion and have other people analyze the data to see if there is anything that was missed. Outside sources work great at taking a different approach to analyzing the data by having a different perception of the information. This can help bring out new information that the manager may have missed the first time. It is important to not make conclusive decisions from the beginning so that there is more room to analyze and look for data that may go against the view that was established. It is also important for managers to not step in with a mindset of being confident with the results right from the beginning because this can lead to managers avoiding data that goes against their views. People who have experience with data decision-making and have been successful in the past should be aware of experiencing overconfidence in the field. Just because managers were successful at formulating the right decision in the past does not mean that mistakes can happen in the future. Managers need to be careful and aware of this. Once the data has been collected and verified to make decisions based on them, managers will need to present the information in a compelling way that can influence the audience in agreeing with what is being offered. Managers should remember to avoid relying on too much data when presenting and instead try being more personal and creating a story that the audience can relate to. Establishing some emotional connections is very important to increase the likelihood of the audience resonating with what you have to offer. Managers should be aware of these methods to succeed in their roles. Establishing a great understanding in correctly interpreting data and presenting them are very important concepts managers will need.
