Resources

Credibility, Evidence, and Footnotes

CONTENT

1.0 Credibility

1.1 Strong Arguments

1.2 Capable Guide

2.0 Evidence

2.1 Personal Observations

2.2 Expert Testimony

2.3 Using Evidence

3.0 Footnotes

1.0 CREDIBILITY

Just below I write about how important details and supportive evidence are to your credibility, but even more important is that you attach those details and that evidence to a strong and logical argument. I'm giving you the argument this time, but when you work on other projects you must first ensure that you have a clear argument that pieces together a set of claims that, if accepted as true, logically lead to a specific and convincing conclusion.

Notice that in the previous sentence the words "if accepted as true" are emphasized. An argument is defined as "sound" if the conclusion logically follows from the set of claims assuming the claims are true. An argument is defined as "valid" if the argument is sound and if all of the argument's claims are, in fact, true. I'm giving you in this text the framework or skeleton for a sound argument. Your task is to ensure that the claims you make as part of the text's prescribed argument are supported by enough evidence and details to create a valid argument.

SUSTAINING AND BUILDING YOUR CREDIBILITY VIDEO (6:13)

1.1 Strong Arguments

The most important way to build your credibility as a judicious, objective, and knowledgeable analyst is to make sure that you have a valid and clearly organized argument. Arguments are comprised of multiple propositions that are supported by theoretical claims and evidence and clarified by definitions and assumptions. It is the analyst's job to make it clear to readers what the propositions are, how they are supported and clarified, and how, if they are true, they logically to a specific conclusion (Figure 1).

Importantly, evidence is used to justify the propositions, and the propositions logically lead to the conclusion. Evidence is not used directly to support or justify conclusions. The conclusion follows logically from the propositions. The relationship between the the propositions and the conclusion is different than the relationship between evidence that supports a proposition. These relationships signify whether an argument is valid and/or sound.

Strong arguments have propositions with two characteristics. The propositions are logically supportive of the conclusion, and the propositions are true. That is the same as saying that the argument is valid and sound, respectively. A deductive argument is valid if, assuming the propositions to be true, one must logically accept that the conclusion is also true. An inductive argument is valid if, assuming the propositions to be true, one might logically accept that the conclusion is also true. In economics, most applied analyses use inductive arguments (i.e., inductive reasoning). That is, most empirical studies in economics try to explain human behavior with inductive arguments. These inductive arguments show what is probably true (i.e., might be true) given a set of propositions that the analyst supports with the best available, but still usually incomplete, evidence.

The prescribed arguments and analytical approaches that are described in this text is valid arguments. If you follow the prescribed analytical approaches, you will arrive at a logical conclusion, provided that you provide sufficient evidence and concrete details to convince readers that each of the argument's propositions are true. If you follow the argument and fill it with convincing evidence, the argument will not only be valid; it will also be sound. 

An argument is sound if the argument is both valid and its propositions are all true. So, it is important first of all that arguments are valid. That is, if we accept the propositions as true, then it follows that the conclusion is (deductive argument) or is probably (inductive argument) true. It is important secondly that arguments are sound, meaning that the propositions are actually true or at least that they are as convincingly true as can be known given the resources available for discovery and investigation. Evidence can come from personal interviews and surveys or from expert testimony published in the secondary literature. Importantly, technical communication uses evidence to support the truth of an argument's propositions but not its conclusions. In other words, technical communicators rarely use evidence (e.g., expert testimony in the secondary literature) to argue that the conclusion is true. Instead, technical communicators use evidence to show that the propositions are true (or probably true) and then let deductive logic (or more often inductive logic) alone justify the conclusion. Often, small bits of well chosen and appropriately referenced evidence (versus a long list of unclear and/or half-related evidence) can provide strong support for an otherwise objectionable proposition and thus diffuse reader/viewer criticism. It's one of the best ways to craft a logical argument and judicious conclusion.

For example, consider an argument with two propositions and a resulting conclusion (also known as a syllogism): 

Proposition 1: People with fair skin sunburn easily.

Proposition 2: Joe is a person with fair skin.

Conclusion: Joe sunburns easily.

In this argument (also called a syllogism), the conclusion logically follows provided the two propositions are, based on the supplied evidence, determined to be true. The first two statements are the propositions. The two propositions should lead logically to the conclusion (i.e., the third statement). For simple arguments, it is easy to know where evidence is needed. Evidence should be used to support the proposition that people with fair skin sunburn easily and that Joe is a person with fair skin. 

Analytical economic arguments are often more complex. Simple arguments are often joined together and coupled with theoretical propositions. As arguments become more complex, it can sometimes be unclear what kind of evidence is needed and how to use it. 

For example, consider a more complex argument: 

1. Declining taxes lead to increased consumer incomes.

2. Increasing stock prices lead to increased consumer incomes.

3. Increasing consumer incomes lead to increase demand for normal goods.

4. Increased demand leads to higher market prices.

5. Thoroughbred race horses are normal goods.

6. Taxes are declining.

7. Stock prices are increasing.

8. Therefore, market prices for Thoroughbred race horses will increase. 

When presented with an argument, you might need to outline it first to identify each proposition clearly. The argument above only includes the propositions and no evidence, so it is already in a simplified outline form. Next, you should check that the argument is valid, meaning that if all of the propositions are assumed to be true, it necessarily follows that the conclusion is true. Next, you should consider what kind and amount of evidence would be needed to show that each proposition is true. Sometimes it is not possible to gather evidence to support a proposition. In those cases, it is necessary to ask your readers and listeners to assume that the proposition is true. Assumptions help simplify arguments, but they also can be seen as weaknesses, particularly if the assumptions are not easily accepted by others. Since arguments are communicated using words, it is important to define words that might have different meanings to different people.

Let's break down this argument. The conclusion (item 8) is that market prices for Thoroughbred race horses will increase. Why is this? This conclusion rests on two theoretical cause-effect relationships. First, consumer incomes increase when taxes decrease (item 1) or stock prices increase (item 2). Second, when consumer incomes increase for normal goods (item 3), demand shifts cause higher market prices (item 4). The specific case simply parallels these two general theoretical cause-effect relationships. Consumer incomes are actually increasing because taxes are actually decreasing (item 6) and stock prices are increasing (item 7). Given this increase in consumer incomes and given that Thoroughbred horses are normal goods, demand will shift higher (4) causing market prices for Thoroughbred horses to increase (item 8).      

The theoretical propositions (i.e., 1, 2, 3, and 4) probably require the least evidence and you might be able to offer these as assumptions. Relatively little evidence is probably required. Whether Thoroughbred race horses are normal goods is an empirical (i.e., not theoretical) question, but few people are likely to disagree if you offer this as an additional assumption. So, more evidence is needed than what is needed for the purely theoretical propositions, but relatively little still. The factual propositions (6 and 7) require strong evidence, if available. These factual propositions are the most critical propositions for this argument. The stronger the evidence is for these two propositions, the more convincing this argument will be.

1.2 Capable Guide

Authors of technical writing should adopt the tone of a reasonable and modest guide. Authors who are reasonable and modest guides will speak and write as one who is leading their audiences patiently and methodically through an analysis, as one who points out along the way where the dead-end trails are, where the best path forward is, and why. 

Reasonable and modest guides do not appear rushed; they do not sound hurried. They do not appear ashamed to admit when time or other resources are insufficient to explore other areas or to give more complete explanations. These authors appear ready and willing to admit the short-comings of the analysis but do not appear to dwell on, take offense, apologize, or make excuses for them. These authors do not seem concerned to convince readers that the selected path is particularly exciting or dull, only that it seems honestly to them like the logical way to go given the available information and the necessary assumptions. This is the tone that technical writers should have. You should avoid biased, hysterical, informal, and hostile tones.

Capable guides recognize when some readers or listeners might disagree or not fully understand something, opting in these instances to insert footnotes to add clarity without unnecessarily interrupting the flow of the main analysis.

In the Paper section of this chapter you will find more specific suggestions about how to craft messages that increase your credibility as a capable guide. Such suggestions include, avoiding problems with figurate language, unfamiliar words, undefined words, distracting tones, and unclear references.

2.0 EVIDENCE

This text gives you detailed instructions for how to conduct a particular kind of analysis, specifically a multiple criteria decision analysis. The text provides you with a detailed outline for each section and lots of prescribed language (i.e., sample text for you to cut-and-paste or alter slightly). The text lays out for you a compelling analytical argument (see "Overview" chapter). That argument includes a number of claims that together logically lead to a specific conclusion. In short, you don't have to worry that the analysis described in this text won't have a strong argument and clear organization. What is left is for you to do is focus on choosing a topic, adapting the topic to the prescribed argument, and then providing  concrete details and supporting evidence to further build your credibility.

FINDING AND USING HIGH QUALITY EVIDENCE VIDEO (6:26)

2.1 Personal Observations

Personal observations are whatever you see, hear, smell, or taste directly. Personal observations are experienced and recorded individually by the analyst. In this way, they differ from observations experienced and recorded by other people (e.g., observations in published sources). 

To be useful in technical analyses, personal observations generally should be representative rather than anecdotal. That means that personal observations should reflect the general situation. If you use personal observations in your market analyses, you should look for and, if possible, include evidence showing that your personal observations are representative. 

For example, suppose you are conducting an analysis of the local farm-direct market for basic, fresh, in-season vegetables. You stop at a vegetable farm with a road-side stand to gather information about this particular producer. You observe fresh corn, tomatoes, and green beans for sale. You observe a very basic covered, open frame structure with vegetables in shallow bins. You observe signs with the prices per unit. You observe a metal box for payment. You see no one at or near the stand. You take several pictures to document what you see. These are all personal observations. However, these personal observations may not be representative of all producers in the market. For evidence, you might visit several other producers in the market area. In your report, you could describe the characteristics that all producers had in common (i.e., the typical characteristics) and then note some of the most significant differences (i.e., provide counter evidence to the claim that all producers are the same). 

2.2 Expert Testimony

One of the most useful kind of evidence for analyses described in this text is expert testimony, particularly primary expert testimony that you solicit and record. Expert testimony can come from primary or secondary sources. Expert testimony is characterized as primary evidence (i.e., from a primary source) when the information is first gathered by you personally as the analyst. Expert testimony is characterized as secondary evidence (i.e., from a secondary source) when the information was first gathered by someone else (e.g., a published interview). For the market analysis assignments described by this text, readers will likely need to conduct one or more interviews with experts in the market. 

Primary expert testimony may include interviews with local and regional business development officers, university researchers, and especially the owners and managers of firms or a party you are helping to solve a problem. While such interviews are not considered human subject research, a different set of ethical guidelines--that for journalists--still applies. Interviews of experts or of representatives of organizations (e.g., business owners) are not considered human subject research if the interviews are limited only to questions about the expert's area of expertise or about the representative's organization. If the interviewer asks personal questions (e.g., about the expert's experience as a business owner), the interview qualifies as a survey, and the investigation qualifies as human subject research. If the business owner, industry expert, or organizational representative tells a personal story (e.g., "These last few months have been challenging from a profit standpoint for me at my shop."), the interviewer must judge whether the information was intended as personal information or as an anecdote, representative of the industry or field as a whole. The latter can be shared; the former probably not. If unsure, ask your instructor.  

1. Journalism ethics. Like human subject research, ethical journalism calls on reporters to minimize harm to sources, subjects, colleagues, and members of the public. Ethical journalism also asks reporters to recognize that private people have a greater right to privacy than public figures who seek power, influence, or attention. A key difference between human subject research ethics and journalism ethics is that the latter expressly asks that reporters identify their sources clearly when possible. The rationale is that the public is entitled to know as much information as possible about alleged experts so the public can make informed judgments about the reliability and motivations of sources. In reality, however, reporters must balance the ethical requirement to minimize harm versus the expectation to reveal sources clearly. In some instances, for example, business owners or producers may report information that, if associated with their firm, could compromise the success of their firm or harm them. If in doubt, ask the source if there is information from your interview that you cannot report and attribute to the interviewee.           

2. How to interview an expert. Like writing and most other skills, learning to interview other people successfully requires dedication and practice. The best interviewers all have one trait in common. They are sincerely curious about the people they interview. Curious people are good listeners. They are respectful of other people's valuable time. They are thankful. 

Here are a few other suggestions for interviewing experts:

• Gather background information about your source. 

If you are interviewing a business owner, search the internet for any information you can find about the particular firm. If you do not have enough background information about the subject or about the particular source, you risk wasting the source's time and embarrassing yourself.

• Ask questions related the source's area of expertise.

Do not ask general information questions that you could find the answers to on your own or though secondary research.

• Make an appointment, if possible. Do not miss an appointment.

If you absolutely must miss an appointment, contact the person in advance. Do not miss an appointment. If you cannot make an appointment, be especially sensitive about the source's needs and about the length of time you take.

• Be clear who you are and your affiliation. 

Take time before you start your interview or when making an appointment to state clearly who you are why you are requesting an interview. Be aware that you will likely receive special consideration if you are a student. But, that also means that you have a duty to represent your school well.

• Be clear what information you want. 

Write a list of questions and related follow-up questions. It is almost always best to ask general questions first to establish a friendly rapport. For example, you might say, "Oh, here is the product that I'm studying. These are so tasty. Have you tried these? They're great." Next, consider asking less specific questions first and then asking for a more precise answer, if possible. For example, you might ask first, "Do you have information you could share about your annual production?" Then listen carefully. Give the source a chance to tell you a precise quantity. If the source does not volunteer that information, you can ask for it specifically. Be aware that sometimes a source may not know the answer to a question. Again, start with the general question and then become more specific. For example, you might ask, "Do you know much about how your levels of production compare to everyone else in the market?" The source likely does not have exact details about his firm's market share. Listen to whatever information is shared and then, if needed, ask, "What would be your best guess [about market share]". Also, avoid specialized terms that your source may not know or that may have different meanings to different people (e.g., "market share", "competitive advantage", "elasticity").

• Be clear what you will do with the information. 

When making an appointment and introducing yourself, be sure to explain what you intend to do with the information you gather. The aim is to help the source feel safe and comfortable sharing information during your interview. If you are a university student who is completing these assignments as part of a class, be sure to mention where you attend school. Busy business owners and managers are sometimes more willing to speak with students who are trying to learn a new skill.

• Keep the interview short. 

Make every minute of your interview count. Listen carefully. Do not interject too quickly. But also, do not waste time. Keep control of the interview and the pace of the questions.

• Thank your source.

2.3 Using Evidence

Don't use evidence to support conclusions. It should be clear that you should not use evidence directly to support the conclusion. For example, suppose your analysis seeks to predict what will happen in the next five years to the equilibrium market price and quantity for Thoroughbred race horses. Suppose you then find expert testimony from a university professor who says, "The equilibrium market prices and quantities for Thoroughbred race horses are expected to increase in next five years." This professor's statement is the conclusion of your argument; however, your conclusion--as an analyst--should not derive from the testimony of such experts. Instead, your conclusion should derive from your argument. Your conclusion should follow logically from the propositions that make up your analysis, not directly from the opinion of an expert. When writers appeal to another expert for direct support of their conclusion, it suggests that the writer's logical analysis is irrelevant and all that matters is the so-called expert's opinion. Do not use expert opinions directly to support your conclusions. Such evidence might be useful to insert when you present your own conclusion and discuss how your conclusion is similar to or different from other recognized experts.

Do not use evidence to justify established theoretical claims (usually). Economic theory in many cases is well established. For example, well-accepted economic theory shows that an increase in consumer incomes shifts the demand curve to the right for normal goods, ceteris paribus. An analyst could, if desired, reference textbooks that provide provide confirmation of this theoretical claim. Or, the analyst could even go back into the academic literature and reference as evidence the original theoretical publications. However, this is not usually expected or recommended. Instead, most applied technical analyses in economics simply assume that the majority of economic theory is common knowledge and, for this reason, do not offer supportive evidence (e.g., references to original theoretical sources). Applied economic analyses usually provide explanations of the theoretical concepts, but they do not usually provide evidence to support undisputed theoretical concepts. 

Do not let evidence substitute for analysis. Just as you do not want someone else's (e.g., an expert's) opinion to be seen as a substitute for your own conclusion, you also do not want to let evidence substitute for your analysis. For example, suppose you find a news article that quotes an expert saying that demand for Thoroughbred race horses is expected to increase. This would not be useful evidence since shifts in demand occur as a result of specific market changes (e.g., changes in consumer incomes). Even if the news article included some explanation for why the expert believes demand will increase, it would still not be relevant as evidence assuming you are the economic analyst. You should avoid substituting someone else's analysis (e.g., their explanations for why demand might shift) for your own analysis (i.e., your explanations and evidence for why demand might shift). 

3.0 FOOTNOTES

Footnotes are an excellent way to provide additional supportive or explanatory information without interrupting the argument or flow of the main text. In other words, footnotes are mostly unobtrusive. Footnotes are also very adaptable. They can be used to provide reference information, additional explanation, or both in a single footnote. Most word processing software (e.g., Microsoft Word and Google Docs) format and renumber footnotes automatically (see here).

FOOTNOTES AND HOW TO USE THEM VIDEO (11:42)

For all of the assignments described in this text, you must put your attribution details (e.g., citations) in footnotes. For consistency, you may not use parenthetical references, endnotes, or a list of sources at the end of your paper (e.g., a bibliography or "Works Cited" page).   

What information goes in the main text and what information goes in footnotes? The general rule with footnotes is that information and details that are essential to the argument (i.e., that all readers need to know) should appear in the main text. Information or details that are not essential to the argument but that merely extend or clarify points in the main text should be put into footnotes.

It might help if you think about what happens when your reader encounters a footnote reference (i.e., the little superscript number) in your main text. At that moment, your reader thinks, "Hmmm, do I have a question at this point?" If not, the reader usually skips the footnote and just keeps reading the main text. But, sometimes a reader thinks, "Oh yes! Now that you mention it, I do have a question." It is at that moment that the reader's eyes move to the bottom (i.e., "foot") of the page where the reader hopes to get more information that answers the reader's question. Your job as the analyst is to anticipate what questions your readers will have and what additional information your readers what to see in each footnote. Sometimes readers want to know the source of some data you have shared and often why you think that is a high-quality source. Other times readers want to know the definition of an unfamiliar term. Or, maybe the reader wants to know that you know that there are technical exceptions to whatever you just wrote. In a footnote, you have a lot of freedom to address these and many other information needs that your readers may have.          

For example, suppose you are analyzing the current local retail market for whole, never-frozen salmon in Central Kentucky. Suppose there are three major producers in this market. If so, all readers of your analysis likely need to know the number and size of the firms in the market, but it is probably not relevant or important that everyone know the names of those three producers. While not necessary, the information would provide useful clarification, particularly for readers who might object because they think there are more or fewer market producers. Providing a list of the known market producers in a footnote allows readers to understand the analytical approach and disagree, if needed. Footnotes also avoid overwhelming typical readers with too much detailed information in the main text. And, concrete details like this (i.e., actually naming the three producers in the market, and you could even provide the neighborhood location of each) will increase your credibility as a precise, thorough, and knowledgeable analyst.  

Let's look at some other examples. Here is one that I adapted from a former student. In 2016, this student was analyzing the current regional farm-level market for commercially-produced raw unprocessed dairy cow milk in Central California from 2016 to 2021. Note how the author uses footnotes to support his analysis without obstructing the flow of the main text. Also, notice how the author's use of details in the footnotes increases the author’s credibility.

In November 2016, a new U.S. president will be elected. All major candidates in this election have had to address increasing concerns from voters about income inequality. One policy proposal is to increase the U.S. federal minimum wage from $7.25 per hour to $15.00 per hour.1 The author of this analysis believes that the probability is high that, after the election, the new President and Congress will enact this policy proposal. The effect of a significantly higher federal minimum wage would likely impact the lowest wage earners most. About half of minimum wage earners in 2014 were age 24 or younger, an important demographic group of milk consumers.2 At the retail level, milk products probably are normal goods, meaning that as consumer incomes increase, retail demand for these products will shift to the right, ceteris paribus.3 Farm-level demand for cow’s milk will respond similarly (shift to the right) since farm-level demand for cow’s milk is derived from retail-level demand for milk, cream, cheese, and other dairy products. If the farm-level demand for cow’s milk shifts to the right as expected, both the equilibrium price and quantity of this product will increase, ceteris paribus. 

-------------------------

1. The U.S. federal minimum wage was last increased in June 2009 to $7.25 per hour. The leading Democratic candidates for President (H. Clinton and B. Sanders) are the strongest proponents of raising the minimum wage and at the time of this analysis (March 2016) polls indicate that both of these Democratic candidates would beat the leading Republican candidate (Trump). Plus, a national survey by Politico in January 2016 showed that two-thirds of U.S. mayors supported raising the minimum wage to $15 per hour. Source: Wofford, Ben and Manuela Tobias. 2016. “A Surprising Number of America’s Mayors Support $15 Minimum Wage.” Politico. (https://www.politico.com/magazine/story/2016/01/mayors-survey-minimum-wage-213563/).

2. According to a 2014 report from the U.S. Bureau of Labor and Statistics (BLS), there were about 3 million workers with wages at or below the federal minimum that together comprised 3.9 percent of all hourly paid workers. Of these, 48.2 percent were age 16 to 24. Source: US Bureau of Labor Statistics. 2015. "Characteristics of minimum wage workers, 2014." (https://www.bls.gov/opub/reports/minimum-wage/archive/characteristics-of-minimum-wage-workers-2014.pdf).

3. A study in 2000 by Lechene found that the income elasticity for fluid whole milk in the UK was -0.17 (i.e., an inferior good). However, the broader category “Milk and Cream” had a slightly positive income elasticity (+0.05) with “Cheese” being clearly positive (+0.23), indicating that dairy-related products are normal goods in general. See Lechene V. 2000. “Income and price elasticities of demand for foods consumed in the home.” In: National Food Survey. London: Department for Environment, Food and Rural Affairs. 2000. 89-109.

Notice too the high quality of the author's supporting and clarifying evidence. The footnoted sources used in the example above are a national newspaper (Politico), a U.S. government site (www.bls.gov), and a peer-reviewed journal article (by V. Lechene). These are all good, strong, credible sources of information. You may not be able to find the highest quality sources in every instance, but it often does not take much more effort to find more reputable sources.

Notice too that this author uses footnotes not only to provide attribution information. He also uses footnotes to help answer the "three questions": (1) What is the evidence, (2) Why does the evidence matter, and (3) How strong is the evidence? Again, it is not enough simply to make a claim and then put a citation to an external resource (e.g., a URL link). Critical readers will want you--the analyst--to answer the three questions for them. Simply providing a URL link is almost never sufficient.

Finally, notice the location of the in-text footnote reference numbers. They appropriately appear immediately after the sentence's ending punctuation. The footnote generally appears at the end of the sentence that contains the information that footnote most appropriately relates to. Sometimes footnotes may appear in the middle of a sentence, particularly if a sentence has more than one footnote. If the footnote relates to an entire paragraph or even an entire section, place the footnote at the end of the first sentence in that section. Footnotes should not be placed in places where they may be confused with an exponent or other kinds of numerals. So, a footnote goes right after the sentence's ending punctuation like this.5 If there is more than one footnote for a single sentence, it can look like this.6,7 You should also consult your style guide (APA, MLA, etc.) to see how to format footnotes.

Here is another example. This student was analyzing a niche market in 2016, namely the current local retail market for gluten-free beer in Lexington, Kentucky from 2016 to 2021. This student defined the market boundaries by tracking the locations of people posting to Yelp.com to include Fayette county and portions of Scott and Woodford counties. This was a creative and mostly useful approach. However, when it came time to assess shift factors, the student lost some credibility because his other evidence was weak. He cited an article published in the Lexington-Herald Leader that showed that the population of Lexington had already (!) increased by 15,000 people from 2010 to 2015. He used this information (about something that already happened in the past) to argue that the demand for gluten-free beer would increase in the future, leading to a corresponding increase in both equilibrium price and quantity, ceteris paribus. This is an illogical use of this evidence. The population increase had already occurred in the past (between 2010 and 2015) and his analysis was looking for changes in population from 2016 to 2021. In other words, if there was a demand shift in this market due to this population change, the shift and its effects on demand were already incorporated into the market and reflected in the market equilibrium in 2016. This student instead needs evidence that shows that the population in the market is expected to change during the period 2016 to 2021 (i.e., the time of the analysis). It would be better if this student provided reasons why he believed the trend would continue from 2016 to 2021. That would work.

There are two other problems with this student's use of evidence. One, the article he cited only talked about increases in population in Lexington/Fayette County, but he defined the market area to include two other counties (Woodford and Scott). So, it would be better if he had looked at all three of these counties and not just one county. A second problem is that the Lexington-Herald Leader is not a very good source for data on population change. The title of the newspaper article cited is: “We're No. 61! Lexington's population surpasses 310,000 in latest U.S. Census Report.” It is not always possible for analysts to look at original data sources, but the U.S. Census data is very easy to find. In this case, search Google for: “US Census Fayette, Kentucky” and the first suggestion provided is a link to the data from the U.S. Census Bureau (https://www.census.gov/quickfacts/fact/table/fayettecountykentucky/PST045219). The point is not that you should never cite a small, local newspaper. You can. But, you should not do it when it takes such a small amount of additional effort (marginal cost) to find and cite the original data source (marginal benefit). The original source would give this author so much more credibility for relatively little extra cost. In this case, the additional benefits (more credibility) are probably worth the additional costs (a little more time and effort).

Consider one more example. This example highlights the benefits of relying more on facts (details) and less on assumptions. This example comes from a student in 2017 who wanted to show that a change in incomes for typical consumers (i.e., so-called Millennials) would shift the demand curve to the right for a certain (normal) good that he was analyzing. This was a good idea. But, the student relied on vague assumptions “that a large majority of this segment will be earning more in five years than they do now” because they will be “graduating college… and will have more disposable income”. This claim is probably true. It is a reasonable assumption. But, it is not very useful or compelling analysis. It would not take much additional effort for this analyst to provide useful information and details about how changes in incomes are correlated with increased levels of education. A simple Internet search of “income and college education” brings up several articles, including one from U.S. News and World Reports dated August 2015 (i.e., current at that time) titled: “Seriously, Go to College”. This is a decent national publication, but its authors do not do much original data analysis. But, in that article, the authors reference the researchers who performed the original analysis (i.e., the St. Louis Federal Reserve). That is an excellent source. A Google search easily led to the report from the St. Louis Federal Reserve. That source showed that lifetime earnings for white men born in 1980 with a college degree are 52% higher than those that only graduate from high school. This is relevant data from a highly credible source that would have helped this student avoid an otherwise vague assumption. These details would have bolstered this student's credibility. And, it didn’t take much additional effort to locate the details in a high-quality source. Again, that is a lot of additional benefit for a relatively little additional cost. In such cases, just do it.