This issue has been on my mind throughout my career, and I thought today was the day to put pen to paper (or at least fingers to keyboard) and jot down a few thoughts.
I first came across this issue when I was recruited as an advisory panelist for a Food and Drug Administration (FDA) Advisory committee meeting (AdCOMM), which was now coming up over 20 years ago and well before I actually went to work for the empire of approval. I was part of a distinguished panel tasked with reviewing a sponsor’s application for a new drug to be licensed in the US. We were asked to vote on three specific questions related to the product’s efficacy and safety, and the FDA would use our discussions and voting to guide their final decision to approve or reject the application.
Anyway, back in the most oversized conference room in the history of government, the old White Oak building on the FDA campus. There were about 25 of us, and the committee was focused on a new analgesic, sponsored by a top pharma firm. The committee consisted mainly of doctors, with one or two card-carrying epidemiologists and an equal number of biostatisticians. Each of us had received our terminal doctoral degrees and had led several large clinical trials; most of us were members of elite Ivy League academic institutions… at least I was at the time! (insert college fight song, “Ten Thousand Men of Harvard…”). The large balance of the panel consisted of MDs, mainly in the analgesic/pain medicine specialties. When the FDA chair opened the meeting, we were asked to go around the table and introduce ourselves, providing our name, academic/professional institution, and our field. I was interested to hear some backgrounds of my fellow panelists, as a significant proportion listed “MPH” behind their names listed on the pre-printed roster/name plates, and a few even had “MSPH” too. I later looked up a few of them. Their degrees were MPH, but they had added the “S”, perhaps unknowingly, that the 19th letter in the alphabet makes a difference when it comes to academic degrees. There were a few legit MSPH degrees in the group, but most of those listed lacked the letter “S”, which turns out to be a reasonably significant omission or commission, as I will come back to later.
As we went around the room, I heard time and again some variant of the following… “I am Dr. Somebody Someone, Chair of the ...blankly blank” But then what was added next surprised me… “I am a clinical epidemiologist,” or “I have an appointment and training in the department of epidemiology and biostatistics”. These phrases were repeated often during the introductions, and it turns out, the scenario has also repeated itself regularly over my career. I have heard these same descriptives used many a time at professional meetings, presentations, and online biographical summaries. And as you can probably guess from my long-winded post here, it bugs me.
The Master of Public Health (MPH) degree elicits much confusion. The PhD is another one, as it seems there is a big difference between the title (Dr.) and the profession---(Doctor). And it is because of this confusion that I have never felt comfortable calling myself “DR”, either verbally or in written form. When my brother-in-law first met me, he was under the impression that a PhD is just someone who is a few courses short of obtaining a medical degree--turns out that he is like many who have a lot of confusion about the degree. People who know me can attest that I am the first one to kindly request that others refer to me by my first or last name. Side note—my good friends call me “NOVAK” and it has become quite the moniker, like other famous single-named people (e.g., Cher, Madonna, Prince, Brittney, Elvis). I feel that I am probably closest to Kramer (from Seinfeld) than anyone else. The reason I fail to acknowledge this title might stem from the (mis)use of the title by non-scientific disciplines, such as the television ministry of Dr. Creflo Dollar or pop psychologist Dr. Phil. But the real reason, or at least one of them, is that back in the day, when you traveled on a federally sponsored ticket, the airlines would always add DR to your name. My old roommate when I first got to Harvard (or shall I say West Somerville--shout out to Davis Square--woop woop) was a real doctor (in training), and to this day, still is—anyway, he was on an international flight when there was an incident, and they needed a real doctor. This generated a long-lasting fear that persists in me today---that somewhere, sometime, somehow, during mid-flight, a flight attendant would flurry back to my coach seat, you know---59F, the dreaded middle seat in the second to the last row in the plane. Hey, at least in my recurrent fear, I’m not sitting in 60C, the one right next to the lavatory door! The conversation goes like this…at least in my mind….
Flight attendant: “Dr. Novak, there is a patient at the front of the plane in 12B who is experiencing chest pains. He’s in bad shape. Is there any way you can come take a look and see what you can do.”
“Dr.” Novak: “Sure, but I’ll need to boot up my laptop first.”
Flight attendant: “Why, I am not sure we have time for that.”
“Dr.” Novak: “Listen, there is no way I will be able to help this person until I have SAS 9.4 fired up and I can run some meta-regressions to quickly summarize the potential effect size ratios from the published literature and account for any potential networked dependencies amongst the correlated outcomes. Don’t you know that angina is highly correlated with syncope, a-fib, b-fib, and any other fib. Plus, I’ll also need to consider potential effect stratifications across potentially confounding relationships with clinical and demographic patient factors. And do not get me started on the possibility of missingness not-at-random in these published studies… this could take hours if I do not get started right away.”
Flight attendant: “Oh, wait, I just got word that there is a dental hygienist on board, and she should be able to handle this… but, thank you for your time. If you need an extra bag of pretzels before we land, I’ll be happy to get that for you, but you still need to pay for adult beverages.”
So, yeah…that is about how that might go, at least in my mind. But I digress back to my initial thought. The MPH degree, as the legend goes, was first conceived and implemented to expand the clinically focused physician workforce to adopt a more community-oriented perspective on the etiology and treatment of disease. The basic public health principle is based on the interaction between the host, disease, and environment model of etiology. In the mid-1990s, there were only a handful, about 20 to 30, accredited schools of public health in the US. By way of comparison, today there are well over 100. But the MPH degree was touted back then as a good way to expand credentialism to physicians who may not necessarily have wanted to stay exclusively in a patient-clinical lane and instead, move into a broader focus of research and development on the industry side or move toward an administrative/public service focus on the other.
One unexpected byproduct of the new MPH degree was the new line of revenue it generated after its introduction. But those training at the elite academic institutions, medical students and residents would be more than willing to incur more debt, a lot of it, for a chance to earn alumni status from an Ivy, grabbing a quick credential that is probably useless in their overall career path. You do not need a degree in public health to do any clinical research. In the words of Matt Damon’s character in “Good Will Hunting” (another great fictional tale of a ‘genius’ Harvard student) “you dropped 150 grand for a degree from Harvard for an education you could have got for $1.50 in late charges at your local public library.” But the antagonist in that scene also had a memorable comeback, and it is especially appropriate here… his retort was...“well, yes, but I will have a degree from Harvard.”
Why do I think the MPH degree is useless? Well, it has to do with the generality of the MPH coursework, which typically takes around 12 to 18 months to complete, though now many programs have added more courses, bringing it up to 24 months. The coursework consists of a general or ‘survey’ class describing the organization of the public health system in the US and abroad. There are a few “applied” methods and data analytics courses. These courses do not require calculators or any mathematical skills, as the instructors spend more time on interpreting the data than on the underlying statistical model assumptions and associated calculations. At the culmination of the program, there is also a required “practicum” consisting of a service project. In contrast, the scientific public health degrees (MSPH, MS) require a thesis, which is a 30+ page research document designed to serve as the backbone for the student’s first peer-reviewed publication as primary author. This is why the “S” is so important in the degree, as it draws a line in the sand between the scientists (MSPH) and the practitioners (MPH). The reasons for separating the MPH programmatic curriculum from the MSPH/MS program within the schools of public health are unknown. Still, one explanation that I’ve heard, and it makes sense, is:
In the early 1990s, all public health students were combined into a single pathway, mainly sharing the same coursework and grading rubric. Over time, the tracts eventually separated into the two paths we see today. The reason for the split was mainly due to differences in the cultural composition of the classrooms. According to multiple faculty members with direct historical knowledge, the physician/MPH students were far less interested in the rote aspects of the field, including the underlying mechanics. Instead, their interests were more toward the application of the knowledge and less in the actual tools used to build the knowledge. The phrases “why do we need to know this” and “I’m never going to use this” turned out to be regular grumblings from the peanut gallery in the MPH camp. This was a source of irritation to the group of students seeking the degree with the “S” in it. The MSPH/MS students, however, were keenly interested in the mechanical parts because they would not only be driving the car but also needed to know how to fix the car should it break down. (Note—this is a nod to two famous Bostonians—Click and Clack---the NPR Tappet Brothers). But, the answer to Harvard’s problem regarding dueling pedagogical interests could only come from Harvard itself---along with the 19th-century doctrine established by Plessy v. Ferguson---the idea of separate but equal. For any constitutional legal buffs reading this…
It shouldn’t surprise anyone that Harvard, like most universities, consists of many schools/colleges, such as law, medicine, public administration, and public health. There is often duplication in the degrees. Take my field---biostats—you may earn your degree(s) from the main campus (BS/MS) or the school of public health, via the MSc or MSPH program. Some disciplines offer specializations within statistics, such as quantitative psychology, computational biology, etc. Regardless of the home department, students take a similar course load, so we are also starting from the same foundational principles. But Harvard was unique compared to its Ivy League sisters because it has an oddly titled school, which is part of the traditional mix of colleagues and programs--- its name is the Harvard Extension School. The Harvard Extension School was developed to offer a Harvard-like education to the working class of Boston and the surrounding towns. The coursework is offered in the evenings and weekends, and is led by instructors who are not Harvard Faculty, but adjuncts who are working professionals. Most of these instructors do not have terminal degrees in their field but do have a Harvard pedigree—most being former students. There was a saying that only a Harvard graduate could teach a Harvard student. We’ll come back to this point a bit later. Nonetheless, the Harvard Extension students have all the perks of being a Harvard graduate, except that their diploma explicitly reads, written in bold type-- “Harvard University Extension”. Oh, there is also that little thing about the entrance requirements being far more lenient than those for students applying to study on the main campus. Thus, the reality of the Harvard Extension program is that the degree and training are different from those students on the main campus. The only equal thing between Extension and the other Harvard Schools/programs is that the tuition money goes to the Harvard Corporation—yes, Corporation. All documents, including pay stubs, list Harvard as a Corporation. I thought that it was funny the first time I saw my pay stub. The way things were run, it just made sense. It was also arrogant, as if we were not ashamed to say the quiet part out loud, “we’re not supposed to make money in academia, but hey, we’re Harvard and you’re not.”
The extension model, as it turns out, was pretty easy to implement within the structure of the master’s level public health training program. And there was also no shortage of candidates who would gladly tack on an extra $50k to their mountain of medical student loans for a chance at an Ivy League degree, and when considering all it takes is a few short months of work, well then, it seemed like a no-brainer. Turns out the MPH was a far easier path to a Harvard degree than the traditional routes, which were and continue to be, extremely competitive. But why is the MPH worth the extra effort if you can legitimately call yourself DR. and you are training at a Harvard residency program? The best way to understand the logic is to understand the status of residency programs in relation to their degree-granting institutions. Again, I’ll draw on my own experience at Harvard.
I do not know the number offhand, but some universities have a combined hospital and medical school, such as (a few of my alma maters, as I have had a few additional ones—the University of Kentucky College of Medicine and the University of Wisconsin-Madison). Many other colleges of medicine connect clinical training through a series of affiliated hospital programs. There is no Harvard Hospital in Massachusetts; only several hospitals run through a separate corporate structure, loosely connected through various residency programs, each with an affiliation to Harvard. Back in the mid to late 90s and early 2000s, Partners HealthCare used to run Mass General Hospital, the Brigham and Women’s Hospital, Mt. Auburn Hospital in Cambridge, Mass Eye and Ear, and a few other specialty hospitals and clinics. Under a separate corporate structure stood Beth Israel, Boston Children’s Hospital, Dana Farber Cancer Center, and Spaulding Rehab. All these organizations were and are still classified as “Harvard Affiliated Hospitals”. Each hospital is responsible for its own internship/residency programs. Yet, these independent hospitals must be attached to an accredited medical school to allow the student/resident to sit for the last parts (Step II and III) of the US Medical Licensing Examination. The larger issue is that being trained at a Harvard-affiliated hospital is NOT the same as being an alumnus of Harvard. When the residency ends, so does your affiliation with the university, and there is no legacy status, no alumni clubs, no Harvard-Yale games. You get a Harvard ID, but it is really to access the libraries, such as Countway next to HSDM and HSPH, the dental and public health programs, respectively. Harvard Medical is in Boston (near Fenway Park), not Cambridge, so it is common for interns and residents never to set foot on Harvard's main campus. My old roommate told me a tale of a colleague he had in residency that went so far as to pay out of pocket for his very own, custom Harvard business cards---logo and all, obtained through a local print shop. I won’t go so far as to say counterfeit, but it may be appropriate because this guy was not ‘officially’ a part of the University. He had to do this on his own and could not use the official Harvard printing office. Harvard is very protective of its media rights, and any logo and printing must go through the official, on-campus university printing office. Attendings at the university-affiliated hospitals do indeed receive an official University title (Clinical Assistant Professor, Clinical Associate/Full Professor). In the bylaws, there is a specific mandate that all faculty must be alumni, so all new faculty recruited from outside the alumni circles are granted alumni status (ad eundem), pro forma, ipso facto, e pluribus unum, and any other Latin phrase that sounds impressive to the uninitiated. But the underlings, from the fellows down to the interns, are not considered part of Harvard's main except for their ID. A side note that residents do have access to the Harvard Faculty Club, but most faculty interactions occur at the alumni club in Boston. My own limited experience at the Harvard faculty club was that most of the patrons at the time were those residents seeking to impress girlfriend/boyfriends and family members, though I must admit that I also brought my father and mother-in-law to be for an Easter brunch my first year on campus. Turns out that the Allen Dershowitz’s and Larry Summers types were likely across the river at the Harvard Alumni club in Boston! (Side note--the Harvard Alumni clubs are located in just about every major metropolis around the globe. I've been to a few, but the Boston Harvard Alumni club is one of the shwankiest.
To return to my initial thoughts---the next question that you might be asking yourself is whether the MPH is really needed if most of the trainees in these prestigious internship and residency programs are filled with those who already possess Harvard’s golden key—the Harvard diploma? Again, we need to turn to a clearer understanding of how internship and residency slots are filled at these upper echelon hospitals to understand the draw of the MPH better.
To answer this, let me relay a story from a friend of mine, who was an MIT undergrad and a graduate of Harvard Medical School. However, she did her internship and residency at a suburban hospital in Philly. She returned to Boston and then to Stanford for her fellowship years. Internship, the year after medical school and before residency in one’s own chosen specialty, is arguably the most demanding time in the training cycle to become a physician. Many physicians will freely admit that their intern year was a blur, and they barely remember much, professionally, from the experience. Most of the time is filled with ‘scut’ work that is mindless and bureaucratic chart-keeping, but is highly critical to patient care. The same statements could also be said about the initial years of residency. It isn’t until you hit the senior/chief resident or move to a fellowship position that your training environment really starts to matter. The story goes that interns and residents have it especially hard at the top echelon hospitals, as the senior leaders (e.g., Chief Residents, Fellows, and Attending Physicians) are not only tasked with seeing cases, but also responsible for generating millions of dollars in research revenue, mostly from clinical trials funded by big pharma or the federal government. And of course, there is the big ego that goes with prestige, making life especially difficult for junior physicians in those training slots.
I recall asking my friend why she didn't stay local and go to Mass General or the Brigham. She said, "Oh, those slots are generally for the kids from Big State U programs." Those internships and residency programs are notoriously difficult, not because they are challenging but because you are working like a “sled dog” –her words, not mine! She eventually did a few fellowships, but those were reserved for the top-shelf dermatology programs, which are among the most competitive specialty programs. She didn’t want to “waste” her time in an internship at one of these ‘meat grinders’ after nearly burning herself out while undergoing the rigors of MIT undergrad and Harvard Med. I recall a conversation we had…and I’m paraphrasing again— “A lot of ‘go-getters’ from Big State U. are in awe of getting an internship at an Ivy-affiliated hospital. What they don’t know is that even though you are being trained at a ‘Harvard Affiliated Hospital’, you are not anywhere near being a Harvard student, except getting the ID and the discount at the COOP.” (note—the Coop is the university bookstore). She said that most of her top-tier classmates went to lesser-known hospitals for an internship, where the pace of life was more manageable than continuing the rat wheel. She very bluntly told me, “I’ve got a Harvard MD, so I really won’t gain much in the intern/residency programs there. But I will come back for a fellowship, as that is where you earn your reputation in academic medicine.”
It turns out that this strategy is relatively common among students graduating from the top medical schools in the US. My Boston roommate graduated from the highly prestigious Dartmouth Medical School, and I would rank him to this day as one of the top five smartest people I’ve ever met in my life. Also, one of the nicest. But he chose to intern at Mt. Auburn Hospital in Cambridge. A hospital with a much lower patient census and less diverse and less complex case mix than those top-tier Hospitals right across the Charles River. Don’t get me wrong, he worked extremely hard, but I recall that he admitted that his intern year could have been so much worse had he chosen a life at Mass General. I remember being a bit surprised by just how much free time he had, relatively speaking. After some years of reflection, now I know why. He eventually landed at Brigham and Women’s, and then fellowships at Boston Children’s, and then finally an attending position at Yale Hospital for anesthesia. Turns out it was a sage decision.
Now, to return to my original question--does the MPH enhance your clinical or research skills? Does it make you a card-carrying “Epidemiologist,” or “Biostatistician,” or “Health Services Researcher”? Let’s be honest, it doesn’t. Granted, the degree provides a credential that you have been exposed to the core ideas involved in the most primitive parts of human subject research. But, as one of those card-carrying researchers, it has taken me many years to hone my craft, and that is the knowledge that only comes with experience. When I was in academic medicine at Brown University, just 45 minutes down the road from Boston, located in Providence, Rhode Island, I had a good friend in my department who was a physician. Most of his time was focused on research, but he contributed one day per week to clinical practice. I think it was to maintain his medical license. He limited his caseload to general internal medicine because, as he put it, ‘I have just lost so much of my clinical intuition, diagnostic skills, and ability to execute my training (originally in neurology) that I would be severely negligent if I chose to move into areas more complex than preventive medicine and general internal medicine. I honestly hope I wouldn’t be put in a life-or-death type of situation for a patient.’
I am not against the MPH degree or the prospects of a physician who is passionate about research. Harvard, Yale, and the other MPH machine-like programs do not resemble the Christmas episode of Oprah’s Favorite Things, where “You get a Harvard degree, and you get a degree, and she gets a degree.” The workload for the MPH can be burdensome, especially when putting in your training dues, which translates to the equivalent of two full-time jobs, simultaneously. But I am 100% against credential inflation---using a degree to lend authority when the degree itself was never intended to substitute for a Master of Science, especially in a field such as epidemiology, biostatistics, economics, or some other experimentally focused field (e.g., experimental psychology, computational biology). Training in the human sciences requires both dedication and time. You must start with the basics and work your way into the more complex scientific problems. The medical training model is based on “see one, do one, teach one.” But scientific research, whether conducted on the bench or the trench, doesn’t lend itself well to quick publications and slam-dunk grant applications. I’m sure there is a similar type of argument that can be made for any “executive” degree, such as the eMBA. You won’t see, at least in my lifetime, an eMD—and that is probably a very good thing for us patients! While I continue my tangential thought, don’t get me started on biostatisticians/data analysts whose education consists entirely of Coursera or EDx course badges.
Many schools have also expanded their revenue targets to include certificate programs ranging in price from a hundred dollars to the 20k and 30k range. For a busy PGY, this might be a more suitable option, but of course, the price tag does not include the price of admission to alumni status.
So, in closing, those are my thoughts on the MPH. As the degree has gained in popularity, I’m sure that many patients really do not understand the degree and how it benefits them as a patient, if at all. But if you are one of those physician-scientists, I’m interested in hearing your thoughts. And should you ever see me on a plane and be in urgent need of biostatistical help for your next grant application, research paper, or FDA submission, you can find me easily in or around seat 59F. I’ll be happy to assist, provided you pay for my adult beverage.
About the author:
Scott Novak, MSc., PhD, is the CEO and Chief Biostatistician at Kingfish Statistics and Data Analytics, Inc., a consulting firm supporting clinical trials and real-world evidence-based studies in the pharmaceutical and medical device sectors. He also holds adjunct appointments in the Departments of Biostatistics (School of Public Health) and Internal Medicine (College of Medicine) at the University of North Carolina at Chapel Hill. He currently lives in the Research Triangle region of North Carolina, but travels back to Cambridge each year, and is relentless in trying to use his expired Harvard ID in hopes of receiving 10% off merch at the Coop in Harvard Square. He has yet to be successful, but remains committed to the cause.
Most firms in business analytics are classified as “For-Profit” because they legally must pay taxes on the revenue brought in. Non-Profit firms (503 C 3) must dedicate a portion of their projects toward advancing science and must pay fee on any commercial (e.g., fee bearing project). Kingfish is non-profit because we have focused nearly 100% of our investment back into the company and our employees. we are often asked what makes us different. It is not just that we have an .org instead of a .com for an e-mail address. Rather, it is our philosophy toward our clients and the industry.
Profit in the consulting field means that the net revenue is largely driven by maximizing project costs and minimizing the labor needed to complete a project.
The goal is to complete each project using the lowest labor, either by capping hours or selecting staff with the lowest wages. These lower wage workers are also the ones who typically also have the least experience and expertise. They also may drive projects costs up because their inexperience adds extra hours to make up for the knowledge gap with more senior staffers.
Over the last decade, we’ve seen a convergence in the contracting sector in a melding of the non-profit and for-profit business philosophies. We’ve seen non-profits holding regular “business development meetings”, call projects “sales”, and entice staff bonuses based on the size of the dollar awarded per project in a fiscal year. These companies often recruit highly accomplished academic researchers from university settings, but then remove any semblance of their intellectual autonomy by mis-attributing the fact that projects are brought in by staff, not by organizations.
Non-profit research organizations are not necessarily open books, per se. We are non-profit because our approach is to offer the most cutting-edge analytic tools to help the small to medium size firms. We do not believe that only the larger, financially stable firms should have access to the suite of data analytics to drive profits. We value confidentiality in working with each client to understand and implement a strategy that works within their resource constraints.
For-profit firms are not necessarily so focused on profits that their judgement is clouded. Again, non-profit research firms are very much like for-profit firms in avoiding drawing controversy or It does, however, take a special type of firm to avoid segregating the management and daily business activities so there is no risk of losing sight of the principals that bring people to work in these settings.
We have chosen the non-profit route because our goal is to worry less about fee maximization and more about the projects we’d like to work on, and how they align with our interests and skillsets. We’ve done the academic research thing, with deeply respected publication and research funding portfolios. We are now at a point in our careers where we want to help support our local business communities in North Carolina and across the United States achieve their greatest potential and impact.
Contact us at: 866-853-7828 or info@kingfishstat.com
The P-Value is probably one of the most mis-interpreted areas of research. Not surprisingly, because our brains naturally try to impose the most simplistic heuristic to efficiently organize information. So yeah, it’s our brain’s fault!
There are two very interrelated concepts surrounding P-Values. The first is the concept is effect size. For example, an Odds Ratio of 2.0 is a stronger effect than an odd’s ratio of 1, which is about 50:50 in the probably of an event occurring in the presence of a covariate (Say males versus females). This is where the confusion lies. Standardized effect sizes are difficult to interpret, largely because of their scale. An effect of moving from 1.0 to 1.5 is not the same as moving from 100 to 150 when looking at something like IQ. Yet on some laboratory tests, going from 6.5 to 7.5 can indicate the difference between pre-diabetes and a diagnosis of diabetes. Instead of taking the time to understand the reference point, it is just easier to say that the Odds Ratio was highly significant, P<.001 and leave it at that!
The true conundrum is that you can have a small effect size highly significant, but a large effect size that is non-significant. The P-Value is really a measure of the strength of the inference about the effect. But it is not a direct measure of the effect’s strength. You can think of it this way, it is the probably of finding a significant effect when the finding is not true if the study were repeated many times.
So how is a P-Value related to a study’s effect size? This part is where a study’s sample size and the homogeneity of the population comes into play. The variance is a measure of dispersion around a measure, say the mean socioeconomic status of a school. The two means can be the same, but the variance around those means can be different. This can occur when one goes into a school that may be relatively homogenous, say a rural school, compared to a suburban school that busses students from an inner city, disadvantaged neighborhood. The means may be the same, but the variability in the population is so much greater in the suburban school than the rural school. As a side note, it is this very reason examine the study’s descriptive statistics, such as mean, variance, standardized variance (standard error/standard deviation) before diving deeper into a study’s findings. Editor’s Note: Do you like the way we wove in our Blog Theme here? Thanks to Mrs. Dombrowski’s 8th grade English composition class!
Here’s where the magic of statistics comes into the fold. Statistical significance can be quickly identified by visually inspecting the standard error bars around 2 separate factors. If they overlap, then there is likely to be no statistically significant effect. If they do not, then we can assume that both come from an independent population and differ significantly from one another.
Editor’s Note (again): Sorry to keep butting in, but there is a hook. Non-overlapping confidence intervals (Standard Errors) are statistically significant. But, there are rare occasions when you can get overlapping Confidence Intervals actually statistically significant. This is because the confidence intervals around the statistical tests are calculated slightly differently than the calculations for the observed mean and variance. Ok, back to the story…
Here’s the trick. It is possible to have the same mean and variance move from statistical non-significance to statistical significance merely by increasing the sample size. This process means we can shorten the width of the standard errors around the test statistic, decreasing the probability value of finding a no effect when one is present. But, if you had a trial of say 30 participants and the p-value was non-significant, you cannot under any circumstances state the same trial would have been significant if the trial recruited another 50 patients. The reason is that you cannot guarantee that the additional 50 patients will respond in the same manner as the first 30!
Editor’s Note: Me again (Jaki). Just also wanted to clarify what Scot was mentioning. Your hypotheses can be stated in many ways. You can hypothesize that the means of two factors in a variable are significantly different (2-tailed), that one is greater than the other (1-tailed), or equivalence (let’s deal with that one later). Often, we hypothesize a two tailed test (means are different) so we can detect an effect in either direction. This is more conservative than a 1 tailed tail test and is the norm in clinical research.
So hopefully you are not confused. But let me introduce one last point. There is a big difference between clinical significance and statistical significance. The latter of which is agnostic to the meaning of the data. In epidemiology, we can see there was a statistically significance change in the population of smokers by .001%, but that may translate to 225,000 persons if we assume the U.S. population to be around 255,000,000. While.001% may not seem big, it is large once you consider the raw numbers. Conversely, you may find that a drug statistically increases body temp by 0.1 degree, which clinically occurs when someone walks up a flight of stairs.
In conclusion, (Do you like the way I set that up!) we hope you now understand what exactly a P-Value represents and why you should avoid discussing it as a measure of an effect size. We like to think about it as how much confidence we can place in our actual finding. I can ask Jaki what the effect size was, and she can tell me it is an Odds Ratio of 2.0, or that adding bacon to my diet will nearly double my risk of heart disease. Wow—that seems like a lot, but then I ask her how confident in the finding, and then she tells me she has either strong confidence (P<.001), or moderately strong confidence (p=.049) in the finding. The level of confidence changes depending upon which covariates (confounders) are added.
The Bottom Line: Our team has tremendous experience in working with both “newbies” to statistics and those who are experience. As former college professors, we enjoy answering all your questions and unlike lawyers, we don’t charge per question! And speaking of school----Van Halen Rulz!
Contact us at: 866-853-7828 or info@kingfishstat.com
Advancements in data analytics have R=revolutionized how information can be used in everyday decision-making. Yet, conventional wisdom still prevails that our instinct is far superior to any data that might be at our disposal.
This conflict between our "gut-intuition" and data-driven decision-making is not new, but can be seen in virtually all aspects of life. Take sports, for example, The famous statistician Bill James created algorithms to estimate the value of certain types of traits among players. Using this data, the team could select players that best matched the statistical profile. For instance, players that had a high on-base percentage were valued over home-run hitters or base stealers. This approach challenged conventional thinking because the baseball management loved to rely on scouting to visually evaluate a player as a 5-tool guy—speed, power, hitting for average, fielding and arm strength. Statistics were less important to the scouting process than the overall gut intuition of the staff. If this story sounds familiar, it was most famously portrayed by Brad Pitt in Moneyball. Because this approach was less popular than the "5-Tool" inventory, a team could get undervalued players who best fit the system. The movie portrayed the Oakland "A's", but several other teams have highly succeeded using this system, such as the Boston Red Sox and the Chicago Cubs. Both teams successfully ended decades-long World Series droughts.
So, what does this have to do with business? Well, many business leaders in organizations also use their "gut" as the key factor in the decision-making process. From purchasing new equipment, to deciding upon new marketing campaigns, gut feeling is king. Who knows your clients better than you, right?
Well... by taking a few simple steps, you can harness the power of your data to Save money, increase revenues, and sky-rocket your client satisfaction. The first step is to work with those who know how to organize data to uncover trends and patterns that can build your economic strategy based on factual information about your organization. Think about this way, we blend your insights with objective data, like a fisherman decides what lake to cast, but then uses electronic depth finders to locate fish and top-shelf equipment to bait and hook the big catch! Our successful strategy is based on these steps:
The Bottom Line: We value your input at every step of the way ensuring that the strengths and limitations of all the data elements are known. We work with you to identify strategic goals for your business, and decide which data will lend the most insight. Our team has the experience and expertise to implement this approach within your business environment, all at a cost and timeframe that is tailored to your needs.
Contact us at: 866-853-7828 or info@kingfishstat.com
