{"id":2002,"date":"2025-02-22T07:00:54","date_gmt":"2025-02-22T07:00:54","guid":{"rendered":"https:\/\/mailitics.com\/index.php\/2025\/02\/22\/do-european-mms-actually-taste-better-than-american-mms\/"},"modified":"2025-02-22T07:00:54","modified_gmt":"2025-02-22T07:00:54","slug":"do-european-mms-actually-taste-better-than-american-mms","status":"publish","type":"post","link":"https:\/\/mailitics.com\/index.php\/2025\/02\/22\/do-european-mms-actually-taste-better-than-american-mms\/","title":{"rendered":"Do European M&amp;Ms Actually Taste Better than American M&amp;Ms?"},"content":{"rendered":"\n<div>Do European M&#038;Ms Actually Taste Better than American M&#038;Ms?<\/div>\n<p> \t<BR><br \/>\n<BR><\/BR><br \/>\n    <!-- no image --><br \/>\n \t<BR><br \/>\n<BR><\/BR><\/p>\n<div>\n<p class=\"wp-block-paragraph\" id=\"97ab\"><em>(Oh, I am the only one who\u2019s been asking this question\u2026? Hm. Well, if you have a minute, please enjoy this exploratory <a href=\"https:\/\/towardsdatascience.com\/tag\/data-analysis\/\" title=\"Data Analysis\">Data Analysis<\/a> \u2014 featuring experimental design, statistics, and interactive visualization \u2014 applied a bit too earnestly to resolve an international debate.)<\/em><\/p>\n<h2 class=\"wp-block-heading\" id=\"0aae\">1. Introduction<\/h2>\n<h3 class=\"wp-block-heading\" id=\"84b3\">1.1 Background and motivation<\/h3>\n<p class=\"wp-block-paragraph\" id=\"27c7\">Chocolate is enjoyed around the world. From ancient practices harvesting organic cacao in the Amazon basin, to chocolatiers sculpting edible art in the mountains of Switzerland, and enormous factories in Hershey, Pennsylvania churning out 70 million kisses per day, the nuanced forms and flavors of chocolate have been integrated into many cultures and their customs. While quality can greatly vary across chocolate products, a well-known, shelf-stable, easily shareable form of chocolate are M&amp;Ms. Readily found by convenience store check-out counters and in hotel vending machines, the brightly colored pellets are a popular treat whose packaging is re-branded to fit nearly any commercializable American holiday.<\/p>\n<p class=\"wp-block-paragraph\" id=\"dd17\">While living in Denmark in 2022, I heard a concerning claim: M&amp;Ms manufactured in Europe taste different, and arguably \u201cbetter,\u201d than M&amp;Ms produced in the United States. While I recognized that fancy European chocolate is indeed quite tasty and often superior to American chocolate, it was unclear to me if the same claim should hold for M&amp;Ms. I learned that many Europeans perceive an \u201cunpleasant\u201d or \u201ctangy\u201d taste in American chocolate, which is largely attributed to\u00a0<a href=\"https:\/\/www.chemistryworld.com\/podcasts\/butyric-acid\/1017662.article\" rel=\"noreferrer noopener\" target=\"_blank\">butyric acid<\/a>, a compound resulting from differences in how milk is treated before incorporation into milk chocolate.<\/p>\n<p class=\"wp-block-paragraph\" id=\"fedd\">But honestly, how much of a difference could this make for M&amp;Ms?\u00a0<em>M&amp;Ms!<\/em>? I imagined M&amp;Ms would retain a relatively processed\/mass-produced\/cheap candy flavor wherever they were manufactured. As the lone American visiting a diverse lab of international scientists pursuing cutting-edge research in biosustainability, I was inspired to break out my data science toolbox and investigate this M&amp;M flavor phenomenon.<\/p>\n<h3 class=\"wp-block-heading\" id=\"d6a8\">1.2 Previous work<\/h3>\n<p class=\"wp-block-paragraph\" id=\"c030\">To quote a European woman, who shall remain anonymous, after she tasted an American M&amp;M while traveling in New York:<\/p>\n<blockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n<p class=\"wp-block-paragraph\" id=\"7a35\">\u201cThey taste so gross. Like vomit. I don\u2019t understand how people can eat this. I threw the rest of the bag away.\u201d<\/p>\n<\/blockquote>\n<p class=\"wp-block-paragraph\" id=\"4a2d\">Vomit? Really? In my experience, children raised in the United States had no qualms about eating M&amp;Ms. Growing up, I was accustomed to bowls of M&amp;Ms strategically placed in high traffic areas around my house to provide readily available sugar. Clearly American M&amp;Ms are edible. But are they significantly different and\/or inferior to their European equivalent?<\/p>\n<p class=\"wp-block-paragraph\" id=\"3cbe\">In response to the anonymous European woman\u2019s scathing report, myself and two other Americans visiting Denmark sampled M&amp;Ms purchased locally in the Lyngby Storcenter F\u00f8tex. We hoped to experience the incredible improvement in M&amp;M flavor that was apparently hidden from us throughout our youths. But curiously, we detected no obvious flavor improvements.<\/p>\n<p class=\"wp-block-paragraph\" id=\"f21b\">Unfortunately, neither preliminary study was able to conduct a side-by-side taste test with proper controls and randomized M&amp;M sampling. Thus, we turn to science.<\/p>\n<h3 class=\"wp-block-heading\" id=\"63dc\">1.3 Study Goals<\/h3>\n<p class=\"wp-block-paragraph\" id=\"48d3\">This study seeks to remedy the previous lack of thoroughness and investigate the following questions:<\/p>\n<ol class=\"wp-block-list\">\n<li class=\"wp-block-list-item\">Is there a\u00a0<strong>global consensus<\/strong>\u00a0that European M&amp;Ms are in fact better than American M&amp;Ms?<\/li>\n<li class=\"wp-block-list-item\">\n<strong>Can Europeans actually detect a difference\u00a0<\/strong>between M&amp;Ms purchased in the US vs in Europe when they don\u2019t know which one they are eating? Or is this a\u00a0<strong>grand, coordinated lie<\/strong>\u00a0amongst Europeans to make Americans feel embarrassed?<\/li>\n<li class=\"wp-block-list-item\">\n<strong>Are Americans actually taste-blind\u00a0<\/strong>to American vs European M&amp;Ms? Or can they taste a difference but simply don\u2019t describe this difference as \u201can improvement\u201d in flavor?<\/li>\n<li class=\"wp-block-list-item\">Can these alleged taste differences be\u00a0<strong>perceived by citizens of other continents<\/strong>? If so, do they find one flavor obviously superior?<\/li>\n<\/ol>\n<h2 class=\"wp-block-heading\" id=\"de98\">2. Methods<\/h2>\n<h3 class=\"wp-block-heading\" id=\"7891\">2.1 Experimental design and data collection<\/h3>\n<p class=\"wp-block-paragraph\" id=\"82df\">Participants were recruited by luring \u2014 er,\u00a0<em>inviting<\/em>\u00a0them to a social gathering (with the promise of free food) that was conveniently co-located with the testing site. Once a participant agreed to pause socializing and join the study, they were positioned at a testing station with a trained experimenter who guided them through the following steps:<\/p>\n<ul class=\"wp-block-list\">\n<li class=\"wp-block-list-item\">Participants sat at a table and received two cups: 1 empty and 1 full of water. With one cup in each hand, the participant was asked to close their eyes, and keep them closed through the remainder of the experiment.<\/li>\n<li class=\"wp-block-list-item\">The experimenter randomly extracted one M&amp;M with a spoon, delivered it to the participant\u2019s empty cup, and the participant was asked to eat the M&amp;M (eyes still closed).<\/li>\n<li class=\"wp-block-list-item\">After eating each M&amp;M, the experimenter collected the taste response by asking the participant to report if they thought the M&amp;M tasted: Especially Good, Especially Bad, or Normal.<\/li>\n<li class=\"wp-block-list-item\">Each participant received a total of 10 M&amp;Ms (5 European, 5 American), one at a time, in a random sequence determined by random.org.<\/li>\n<li class=\"wp-block-list-item\">Between eating each M&amp;M, the participant was asked to take a sip of water to help \u201ccleanse their palate.\u201d<\/li>\n<li class=\"wp-block-list-item\">\n<strong>Data collected<\/strong>: for each participant, the experimenter recorded the participant\u2019s\u00a0<strong>continent of origin (<\/strong>if this was ambiguous, the participant was asked to list the continent on which they have the strongest memories of eating candy as a child). For each of the 10 M&amp;Ms delivered, the experimenter recorded the\u00a0<strong>M&amp;M origin<\/strong>\u00a0(\u201cDenmark\u201d or \u201cUSA\u201d), the\u00a0<strong>M&amp;M color,<\/strong>\u00a0and the participant\u2019s\u00a0<strong>taste response<\/strong>. Experimenters were also encouraged to jot down any amusing phrases uttered by the participant during the test, recorded under\u00a0<strong>notes\u00a0<\/strong>(data available\u00a0<a href=\"https:\/\/github.com\/erinhwilson\/mnm-taste-test\/tree\/main\/data\" target=\"_blank\" rel=\"noreferrer noopener\">here<\/a>).<\/li>\n<\/ul>\n<h3 class=\"wp-block-heading\" id=\"5d84\">2.2 Sourcing materials and recruiting participants<\/h3>\n<p class=\"wp-block-paragraph\" id=\"bef7\">Two bags of M&amp;Ms were purchased for this study. The American-sourced M&amp;Ms (\u201cUSA M&amp;M\u201d) were acquired at the SFO airport and delivered by the author\u2019s parents, who visited her in Denmark. The European-sourced M&amp;Ms (\u201cDenmark M&amp;M\u201d) were purchased at a local F\u00f8tex grocery store in Lyngby, a little north of Copenhagen.<\/p>\n<p class=\"wp-block-paragraph\" id=\"2211\">Experiments were conducted at two main time points. The first 14 participants were tested in Lyngby, Denmark in August 2022. They mostly consisted of friends and housemates the author met at the Novo Nordisk Foundation Center for Biosustainability at the Technical University of Denmark (DTU) who came to a \u201cgoing away party\u201d into which the experimental procedure was inserted. A few additional friends and family who visited Denmark were also tested during their travels (e.g. on the train).<\/p>\n<p class=\"wp-block-paragraph\" id=\"9e6d\">The remaining 37 participants were tested in Seattle, WA, USA in October 2022, primarily during a \u201cTGIF happy hour\u201d hosted by graduate students in the computer science PhD program at the University of Washington. This second batch mostly consisted of students and staff of the Paul. G. Allen School of Computer Science &amp; Engineering (UW CSE) who responded to the weekly Friday summoning to the Allen Center atrium for free snacks and drinks.<\/p>\n<figure class=\"wp-block-image size-large\"><img data-recalc-dims=\"1\" data-dominant-color=\"dae3eb\" data-has-transparency=\"false\" style=\"--dominant-color: #dae3eb;\" loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"519\" src=\"https:\/\/i0.wp.com\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/0_Kx704B8WXguz68lU-1024x519.webp?resize=1024%2C519&#038;ssl=1\" alt=\"\" class=\"wp-image-598318 not-transparent\" srcset=\"https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/0_Kx704B8WXguz68lU-1024x519.webp 1024w, https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/0_Kx704B8WXguz68lU-300x152.webp 300w, https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/0_Kx704B8WXguz68lU-768x389.webp 768w, https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/0_Kx704B8WXguz68lU.webp 1400w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\"><figcaption class=\"wp-element-caption\">Figure 1. Distribution of participants recruited to the study. In the first sampling event in Lyngby, participants primarily hailed from North America and Europe, and a few additionally came from Asia, South America, or Australia. Our second sampling event in Seattle greatly increased participants, primarily from North America and Asia, and a few more from Europe. Neither event recruited participants from Africa. Figure made with Altair.<\/figcaption><\/figure>\n<p class=\"wp-block-paragraph\" id=\"ebc5\">While this study set out to analyze global trends, unfortunately data was only collected from 51 participants the author was able to lure to the study sites and is not well-balanced nor representative of the 6 inhabited continents of Earth (Figure 1). We hope to improve our recruitment tactics in future work. For now, our analytical power with this dataset is limited to response trends for individuals from North America, Europe, and Asia, highly biased by subcommunities the author happened to engage with in late 2022.<\/p>\n<h2 class=\"wp-block-heading\" id=\"0c04\">2.3 Risks<\/h2>\n<p class=\"wp-block-paragraph\" id=\"69ff\">While we did not acquire formal approval for experimentation with human test subjects, there were minor risks associated with this experiment: participants were warned that they may be subjected to increased levels of sugar and possible \u201cunpleasant flavors\u201d as a result of participating in this study. No other risks were anticipated.<\/p>\n<p class=\"wp-block-paragraph\" id=\"4558\">After the experiment however, we unfortunately observed several cases of deflated pride when a participant learned their taste response was skewed more positively towards the M&amp;M type they were not expecting. This pride deflation seemed most severe among European participants who learned their own or their fianc\u00e9\u2019s preference skewed towards USA M&amp;Ms, though this was not quantitatively measured and cannot be confirmed beyond anecdotal evidence.<\/p>\n<h2 class=\"wp-block-heading\" id=\"c29a\">3. Results &amp; Discussion<\/h2>\n<h3 class=\"wp-block-heading\" id=\"598a\">3.1 Overall response to \u201cUSA M&amp;Ms\u201d vs \u201cDenmark M&amp;Ms\u201d<\/h3>\n<h4 class=\"wp-block-heading\" id=\"59e5\"><strong>3.1.1 Categorical response analysis \u2014 entire dataset<\/strong><\/h4>\n<p class=\"wp-block-paragraph\" id=\"0825\">In our first analysis, we count the total number of \u201cBad\u201d, \u201cNormal\u201d, and \u201cGood\u201d taste responses and report the percentage of each response received by each M&amp;M type. M&amp;Ms from Denmark more frequently received \u201cGood\u201d responses than USA M&amp;Ms but also more frequently received \u201cBad\u201d responses. M&amp;Ms from the USA were most frequently reported to taste \u201cNormal\u201d (Figure 2). This may result from the elevated number of participants hailing from North America, where the USA M&amp;M is the default and thus more \u201cNormal,\u201d while the Denmark M&amp;M was more often perceived as better or worse than the baseline.<\/p>\n<p><iframe loading=\"lazy\" class=\"wp-block-html\" width=\"100%\" height=\"432\" frameborder=\"0\" src=\"https:\/\/observablehq.com\/embed\/48936dcc6380d2ba?cells=viewof+chart2\"><\/iframe><\/p>\n<p class=\"wp-block-paragraph\"><sup>Figure 2. Qualitative taste response distribution across the whole dataset. The percentage of taste responses for \u201cBad\u201d, \u201cNormal\u201d or \u201cGood\u201d was calculated for each type of M&amp;M. Figure made with Altair.<\/sup><\/p>\n<p class=\"wp-block-paragraph\" id=\"8829\">Now let\u2019s break out some <a href=\"https:\/\/towardsdatascience.com\/tag\/statistics\/\" title=\"Statistics\">Statistics<\/a>, such as a\u00a0<em>chi<\/em>-squared (X2) test to compare our observed distributions of categorical taste responses. Using the scipy.stats\u00a0<a href=\"https:\/\/docs.scipy.org\/doc\/scipy\/reference\/generated\/scipy.stats.chi2_contingency.html\" rel=\"noreferrer noopener\" target=\"_blank\">chi2_contingency<\/a>\u00a0function, we built contingency tables of the observed counts of \u201cGood,\u201d \u201cNormal,\u201d and \u201cBad\u201d responses to each M&amp;M type. Using the X2 test to evaluate the null hypothesis that there is no difference between the two M&amp;Ms, we found the\u00a0<em>p<\/em>-value for the test statistic to be 0.0185, which is significant at the common\u00a0<em>p<\/em>-value cut off of 0.05, but not at 0.01. So a solid \u201cmaybe,\u201d depending on whether you\u2019d like this result to be significant or not.<\/p>\n<h4 class=\"wp-block-heading\" id=\"85c9\"><strong>3.1.2 Quantitative response analysis \u2014 entire dataset.<\/strong><\/h4>\n<p class=\"wp-block-paragraph\" id=\"e788\">The X2 test helps evaluate if there is a difference in categorical responses, but next, we want to determine a relative taste\u00a0<em>ranking<\/em>\u00a0between the two M&amp;M types. To do this, we converted taste responses to a quantitative distribution and calculated a\u00a0<strong>taste score.\u00a0<\/strong>Briefly, \u201cBad\u201d = 1, \u201cNormal\u201d = 2, \u201cGood\u201d = 3. For each participant, we averaged the taste scores across the 5 M&amp;Ms they tasted of each type, maintaining separate taste scores for each M&amp;M type.<\/p>\n<figure class=\"wp-block-image size-large\"><img data-recalc-dims=\"1\" data-dominant-color=\"f1e7eb\" data-has-transparency=\"false\" style=\"--dominant-color: #f1e7eb;\" loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"683\" src=\"https:\/\/i0.wp.com\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/0_0GnxV2PXKBpQkGob-1024x683.webp?resize=1024%2C683&#038;ssl=1\" alt=\"\" class=\"wp-image-598320 not-transparent\" srcset=\"https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/0_0GnxV2PXKBpQkGob-1024x683.webp 1024w, https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/0_0GnxV2PXKBpQkGob-300x200.webp 300w, https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/0_0GnxV2PXKBpQkGob-768x512.webp 768w, https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/0_0GnxV2PXKBpQkGob.webp 1200w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\"><figcaption class=\"wp-element-caption\">Figure 3. Quantitative taste score distributions across the whole dataset. Kernel density estimation of the average taste score calculated for each participant for each M&amp;M type. Figure made with Seaborn.<\/figcaption><\/figure>\n<p class=\"wp-block-paragraph\" id=\"bf38\">With the average taste score for each M&amp;M type in hand, we turn to scipy.stats\u00a0<a href=\"https:\/\/docs.scipy.org\/doc\/scipy\/reference\/generated\/scipy.stats.ttest_ind.html\" rel=\"noreferrer noopener\" target=\"_blank\">ttest_ind<\/a>\u00a0(\u201cT-test\u201d) to evaluate if the means of the USA and Denmark M&amp;M taste scores are different (the null hypothesis being that the means are identical). If the means are significantly different, it would provide evidence that one M&amp;M is perceived as significantly tastier than the other.<\/p>\n<p class=\"wp-block-paragraph\" id=\"b962\">We found the average taste scores for USA M&amp;Ms and Denmark M&amp;Ms to be quite close (Figure 3), and not significantly different (T-test:\u00a0<em>p\u00a0<\/em>= 0.721). Thus, across all participants, we do not observe a difference between the perceived taste of the two M&amp;M types (or if you enjoy parsing triple negatives: \u201cwe\u00a0<em>cannot<\/em>\u00a0<strong>reject<\/strong>\u00a0the null hypothesis that there is\u00a0<em>not<\/em>\u00a0a difference\u201d).<\/p>\n<p class=\"wp-block-paragraph\" id=\"d741\">But does this change if we separate participants by continent of origin?<\/p>\n<h3 class=\"wp-block-heading\" id=\"6fd9\">3.2 Continent-specific responses to \u201cUSA M&amp;Ms\u201d vs \u201cDenmark M&amp;Ms\u201d<\/h3>\n<p class=\"wp-block-paragraph\" id=\"4863\">We repeated the above X2 and T-test analyses after grouping participants by their continents of origin. The Australia and South America groups were combined as a minimal attempt to preserve data privacy. Due to the relatively small sample size of even the combined Australia\/South America group (<em>n<\/em>=3), we will refrain from analyzing trends for this group but include the data in several figures for completeness and enjoyment of the participants who may eventually read this.<\/p>\n<h4 class=\"wp-block-heading\" id=\"44d7\"><strong>3.2.1 Categorical response analysis \u2014 by continent<\/strong><\/h4>\n<p class=\"wp-block-paragraph\" id=\"eeb0\">In Figure 4, we display both the taste response counts (upper panel,\u00a0<em>note the interactive legend<\/em>) and the response percentages (lower panel) for each continent group. Both North America and Asia follow a similar trend to the whole population dataset: participants report Denmark M&amp;Ms as \u201cGood\u201d more frequently than USA M&amp;Ms, but also report Denmark M&amp;Ms as \u201cBad\u201d more frequently. USA M&amp;Ms were most frequently reported as \u201cNormal\u201d (Figure 4).<\/p>\n<p class=\"wp-block-paragraph\" id=\"bd13\">On the contrary, European participants report USA M&amp;Ms as \u201cBad\u201d nearly 50% of the time and \u201cGood\u201d only 18% of the time, which is the most negative and least positive response pattern, respectively (when excluding the under-sampled Australia\/South America group).<\/p>\n<p><iframe loading=\"lazy\" class=\"wp-block-html\" width=\"100%\" height=\"590\" frameborder=\"0\" src=\"https:\/\/observablehq.com\/embed\/48936dcc6380d2ba?cells=viewof+chart4\"><\/iframe><\/p>\n<p class=\"wp-block-paragraph\"><sup>Figure 4. Qualitative taste response distribution by continent. Upper panel: counts of taste responses \u2014 click the legend to interactively filter! Lower panel: percentage of taste responses for each type of M&amp;M. Figure made with Altair.<\/sup><\/p>\n<p class=\"wp-block-paragraph\" id=\"245f\">This appeared striking in bar chart form, however only North America had a significant X2\u00a0<em>p<\/em>-value (<strong><em>p<\/em>\u00a0= 0.0058<\/strong>) when evaluating each continent\u2019s difference in taste response profile between the two M&amp;M types. The European\u00a0<em>p<\/em>-value is perhaps \u201capproaching significance\u201d in some circles, but we\u2019re about to accumulate several more hypothesis tests and should be mindful of multiple hypothesis testing (Table 1). A false positive result here would be devastating.<\/p>\n<figure class=\"wp-block-image size-large\"><img data-recalc-dims=\"1\" data-dominant-color=\"e5e5e5\" data-has-transparency=\"false\" style=\"--dominant-color: #e5e5e5;\" loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"206\" src=\"https:\/\/i0.wp.com\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/0_88VJB8-GJWx0L1XF-1024x206.webp?resize=1024%2C206&#038;ssl=1\" alt=\"\" class=\"wp-image-598323 not-transparent\" srcset=\"https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/0_88VJB8-GJWx0L1XF-1024x206.webp 1024w, https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/0_88VJB8-GJWx0L1XF-300x60.webp 300w, https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/0_88VJB8-GJWx0L1XF-768x155.webp 768w, https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/0_88VJB8-GJWx0L1XF.webp 1400w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\"><\/figure>\n<p class=\"wp-block-paragraph\" id=\"a451\">When comparing the taste response profiles between two continents for the same M&amp;M type, there are a couple interesting notes. First, we observed no major taste discrepancies between all pairs of continents when evaluating Denmark M&amp;Ms \u2014 the world seems generally consistent in their range of feelings about M&amp;Ms sourced from Europe (right column X2\u00a0<em>p<\/em>-values, Table 2). To visualize this comparison more easily, we reorganize the bars in Figure 4 to group them by M&amp;M type (Figure 5).<\/p>\n<p><iframe loading=\"lazy\" class=\"wp-block-html\" width=\"100%\" height=\"386\" frameborder=\"0\" src=\"https:\/\/observablehq.com\/embed\/48936dcc6380d2ba?cells=viewof+chart5\"><\/iframe><\/p>\n<p class=\"wp-block-paragraph\"><sup>Figure 5. Qualitative taste response distribution by M&amp;M type, reported as percentages. (Same data as Figure 4 but re-arranged). Figure made with Altair.<\/sup><\/p>\n<p class=\"wp-block-paragraph\" id=\"6da0\">However, when comparing continents to each other in response to USA M&amp;Ms, we see larger discrepancies. We found one pairing to be significantly different: European and North American participants evaluated USA M&amp;Ms very differently (<strong><em>p<\/em>\u00a0= 0.000007<\/strong>) (Table 2). It seems very unlikely that this observed difference is by random chance (left column, Table 2).<\/p>\n<figure class=\"wp-block-image size-large\"><img data-recalc-dims=\"1\" data-dominant-color=\"e5e5e5\" data-has-transparency=\"false\" style=\"--dominant-color: #e5e5e5;\" loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"404\" src=\"https:\/\/i0.wp.com\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/1_pMPYxYqI3cOsGEBfZg6YBQ-1024x404.webp?resize=1024%2C404&#038;ssl=1\" alt=\"\" class=\"wp-image-598325 not-transparent\" srcset=\"https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/1_pMPYxYqI3cOsGEBfZg6YBQ-1024x404.webp 1024w, https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/1_pMPYxYqI3cOsGEBfZg6YBQ-300x118.webp 300w, https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/1_pMPYxYqI3cOsGEBfZg6YBQ-768x303.webp 768w, https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/1_pMPYxYqI3cOsGEBfZg6YBQ.webp 1400w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\"><\/figure>\n<h4 class=\"wp-block-heading\" id=\"eb61\"><strong>3.2.2 Quantitative response analysis \u2014 by continent<\/strong><\/h4>\n<p class=\"wp-block-paragraph\" id=\"2b7e\">We again convert the categorical profiles to quantitative distributions to assess continents\u2019 relative preference of M&amp;M types. For North America, we see that the taste score means of the two M&amp;M types are actually quite similar, but there is a higher density around \u201cNormal\u201d scores for USA M&amp;Ms (Figure 6A). The European distributions maintain a bit more of a separation in their means (though not quite significantly so), with USA M&amp;Ms scoring lower (Figure 6B). The taste score distributions of Asian participants is most similar (Figure 6C).<\/p>\n<p class=\"wp-block-paragraph\" id=\"d404\">Reorienting to compare the quantitative means between continents\u2019 taste scores for the same M&amp;M type, only the comparison between North American and European participants on USA M&amp;Ms is significantly different based on a T-test (<strong><em>p<\/em>\u00a0= 0.001<\/strong>) (Figure 6D), though now we\u00a0<em>really<\/em>\u00a0are in danger of multiple hypothesis testing! Be cautious if you are taking this analysis at all seriously.<\/p>\n<figure class=\"wp-block-image size-large\"><img data-recalc-dims=\"1\" data-dominant-color=\"edece9\" data-has-transparency=\"false\" style=\"--dominant-color: #edece9;\" loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"625\" src=\"https:\/\/i0.wp.com\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/1_Kqss_it5JWeVwVjOjGoobg-1024x625.webp?resize=1024%2C625&#038;ssl=1\" alt=\"\" class=\"wp-image-598327 not-transparent\" srcset=\"https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/1_Kqss_it5JWeVwVjOjGoobg-1024x625.webp 1024w, https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/1_Kqss_it5JWeVwVjOjGoobg-300x183.webp 300w, https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/1_Kqss_it5JWeVwVjOjGoobg-768x468.webp 768w, https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/1_Kqss_it5JWeVwVjOjGoobg.webp 1400w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\"><figcaption class=\"wp-element-caption\">Figure 6. Quantitative taste score distributions by continent. Kernel density estimation of the average taste score calculated for each each continent for each M&amp;M type.\u00a0<strong>A.<\/strong>\u00a0Comparison of North America responses to each M&amp;M.\u00a0<strong>B.\u00a0<\/strong>Comparison of Europe responses to each M&amp;M.\u00a0<strong>C.<\/strong>\u00a0Comparison of Asia responses to each M&amp;M.\u00a0<strong>D.<\/strong>\u00a0Comparison of continents for USA M&amp;Ms.\u00a0<strong>E.\u00a0<\/strong>Comparison of continents for Denmark M&amp;Ms. Figure made with Seaborn.<\/figcaption><\/figure>\n<p class=\"wp-block-paragraph\" id=\"8c67\">At this point, I feel myself considering that maybe Europeans are not just making this up. I\u2019m not saying it\u2019s as dramatic as some of them claim, but perhaps a difference does indeed exist\u2026 To some degree, North American participants also perceive a difference, but the evaluation of Europe-sourced M&amp;Ms is not consistently positive or negative.<\/p>\n<h3 class=\"wp-block-heading\" id=\"0244\">3.3 M&amp;M taste alignment chart<\/h3>\n<p class=\"wp-block-paragraph\" id=\"f914\">In our analyses thus far, we did not account for the baseline differences in M&amp;M appreciation between participants. For example, say Person 1 scored all Denmark M&amp;Ms as \u201cGood\u201d and all USA M&amp;Ms as \u201cNormal\u201d, while Person 2 scored all Denmark M&amp;Ms as \u201cNormal\u201d and all USA M&amp;Ms as \u201cBad.\u201d They would have the same relative preference for Denmark M&amp;Ms over USA M&amp;Ms, but Person 2 perhaps just does not enjoy M&amp;Ms as much as Person 1, and the relative preference signal is muddled by averaging the raw scores.<\/p>\n<p class=\"wp-block-paragraph\" id=\"fe99\">Inspired by the Lawful\/Chaotic x Good\/Evil alignment chart used in tabletop role playing games like Dungeons &amp; Dragons\u00a9<img data-recalc-dims=\"1\" decoding=\"async\" src=\"https:\/\/i0.wp.com\/s.w.org\/images\/core\/emoji\/15.0.3\/72x72\/2122.png?ssl=1\" alt=\"\u2122\" class=\"wp-smiley\" style=\"height: 1em; max-height: 1em;\">, in Figure 7, we establish an M&amp;M alignment chart to help determine the distribution of participants across M&amp;M enjoyment classes.<\/p>\n<figure class=\"wp-block-image size-large\"><img data-recalc-dims=\"1\" data-dominant-color=\"efeef1\" data-has-transparency=\"false\" style=\"--dominant-color: #efeef1;\" loading=\"lazy\" decoding=\"async\" width=\"986\" height=\"1024\" src=\"https:\/\/i0.wp.com\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/1_ZcqEqqPfxTOs4z4mTET3ng-986x1024.webp?resize=986%2C1024&#038;ssl=1\" alt=\"\" class=\"wp-image-598328 not-transparent\" srcset=\"https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/1_ZcqEqqPfxTOs4z4mTET3ng-986x1024.webp 986w, https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/1_ZcqEqqPfxTOs4z4mTET3ng-289x300.webp 289w, https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/1_ZcqEqqPfxTOs4z4mTET3ng-768x798.webp 768w, https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/1_ZcqEqqPfxTOs4z4mTET3ng.webp 1337w\" sizes=\"auto, (max-width: 986px) 100vw, 986px\"><figcaption class=\"wp-element-caption\">Figure 7. M&amp;M enjoyment alignment chart. The x-axis represents a participant\u2019s average taste score for USA M&amp;Ms; the y-axis is a participant\u2019s average taste score for Denmark M&amp;Ms. Figure made with Altair.<\/figcaption><\/figure>\n<p class=\"wp-block-paragraph\" id=\"cdc4\">Notably, the upper right quadrant where both M&amp;M types are perceived as \u201cGood\u201d to \u201cNormal\u201d is mostly occupied by North American participants and a few Asian participants. All European participants land in the left half of the figure where USA M&amp;Ms are \u201cNormal\u201d to \u201cBad\u201d, but Europeans are somewhat split between the upper and lower halves, where perceptions of Denmark M&amp;Ms range from \u201cGood\u201d to \u201cBad.\u201d<\/p>\n<p class=\"wp-block-paragraph\" id=\"95fd\">An interactive version of Figure 7 is provided below for the reader to explore the counts of various M&amp;M alignment regions.<\/p>\n<p><iframe loading=\"lazy\" class=\"wp-block-html\" width=\"100%\" height=\"443\" frameborder=\"0\" src=\"https:\/\/observablehq.com\/embed\/48936dcc6380d2ba?cells=viewof+chart7\"><\/iframe><\/p>\n<p class=\"wp-block-paragraph\"><sup>Figure 7 (interactive): click and brush your mouse over the scatter plot to see the counts of continents in different M&amp;M enjoyment regions. Figure made with Altair.<\/sup><\/p>\n<h3 class=\"wp-block-heading\" id=\"c080\">3.4 Participant taste response ratio<\/h3>\n<p class=\"wp-block-paragraph\" id=\"e443\">Next, to factor out baseline M&amp;M enjoyment and focus on participants\u2019 relative preference between the two M&amp;M types, we took the log ratio of each person\u2019s\u00a0<strong>USA M&amp;M taste score average<\/strong>\u00a0divided by their\u00a0<strong>Denmark M&amp;M taste score average<\/strong>.<\/p>\n<figure class=\"wp-block-image size-large\"><img data-recalc-dims=\"1\" data-dominant-color=\"f4f4f4\" data-has-transparency=\"false\" style=\"--dominant-color: #f4f4f4;\" loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"460\" src=\"https:\/\/i0.wp.com\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/0_ukUApi5cHzOhMPV8-1024x460.webp?resize=1024%2C460&#038;ssl=1\" alt=\"\" class=\"wp-image-598331 not-transparent\" srcset=\"https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/0_ukUApi5cHzOhMPV8-1024x460.webp 1024w, https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/0_ukUApi5cHzOhMPV8-300x135.webp 300w, https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/0_ukUApi5cHzOhMPV8-768x345.webp 768w, https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/0_ukUApi5cHzOhMPV8-1536x690.webp 1536w, https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/0_ukUApi5cHzOhMPV8.webp 1584w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\"><figcaption class=\"wp-element-caption\">Equation 1: Equation to calculate each participant\u2019s overall M&amp;M preference ratio.<\/figcaption><\/figure>\n<p class=\"wp-block-paragraph\" id=\"8bfa\">As such, positive scores indicate a preference towards USA M&amp;Ms while negative scores indicate a preference towards Denmark M&amp;Ms.<\/p>\n<p class=\"wp-block-paragraph\" id=\"e3c5\">On average, European participants had the strongest preference towards Denmark M&amp;Ms, with Asians also exhibiting a slight preference towards Denmark M&amp;Ms (Figure 8). To the two Europeans who exhibited deflated pride upon learning their slight preference towards USA M&amp;Ms, fear not: you did not think USA M&amp;Ms were \u201cGood,\u201d but simply ranked them as less bad than Denmark M&amp;Ms (see participant_id 4 and 17 in the interactive version of Figure 7). If you assert that M&amp;Ms are a bad American invention not worth replicating and return to consuming artisanal European chocolate, your honor can likely be restored.<\/p>\n<figure class=\"wp-block-image size-large\"><img data-recalc-dims=\"1\" data-dominant-color=\"e2e3de\" data-has-transparency=\"false\" style=\"--dominant-color: #e2e3de;\" loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"780\" src=\"https:\/\/i0.wp.com\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/0_OIJenV8U5Dddt_X8-1024x780.webp?resize=1024%2C780&#038;ssl=1\" alt=\"\" class=\"wp-image-598332 not-transparent\" srcset=\"https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/0_OIJenV8U5Dddt_X8-1024x780.webp 1024w, https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/0_OIJenV8U5Dddt_X8-300x228.webp 300w, https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/0_OIJenV8U5Dddt_X8-768x585.webp 768w, https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/0_OIJenV8U5Dddt_X8.webp 1400w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\"><figcaption class=\"wp-element-caption\">Figure 8. Distribution of participant M&amp;M preference ratios by continent. Preference ratios are calculated as in Equation 1. Positive numbers indicate a relative preference for USA M&amp;Ms, while negative indicate a relative preference for Denmark M&amp;Ms. Figure made with Seaborn.<\/figcaption><\/figure>\n<p class=\"wp-block-paragraph\" id=\"f01a\">North American participants are pretty split in their preference ratios: some fall quite neutrally around 0, others strongly prefer the familiar USA M&amp;M, while a handful moderately prefer Denmark M&amp;Ms. Anecdotally, North Americans who learned their preference skewed towards European M&amp;Ms displayed signals of inflated pride, as if their results signaled posh refinement.<\/p>\n<p class=\"wp-block-paragraph\" id=\"f8d5\">Overall, a T-test comparing the distributions of M&amp;M preference ratios shows a possibly significant difference in the means between European and North American participants (<strong><em>p<\/em>\u00a0= 0.049<\/strong>), but come on, this is like the 20th p-value I\u2019ve reported \u2014 this one is probably too close to call.<\/p>\n<h3 class=\"wp-block-heading\" id=\"afe8\">3.5 Taste inconsistency and \u201cPerfect Classifiers\u201d<\/h3>\n<p class=\"wp-block-paragraph\" id=\"389f\">For each participant, we assessed their taste score consistency by averaging the standard deviations of their responses to each M&amp;M type, and plotting that against their preference ratio (Figure 9).<\/p>\n<p><iframe loading=\"lazy\" class=\"wp-block-html\" width=\"100%\" height=\"457\" frameborder=\"0\" src=\"https:\/\/observablehq.com\/embed\/48936dcc6380d2ba?cells=viewof+chart9\"><\/iframe><\/p>\n<p class=\"wp-block-paragraph\"><sup>Figure 9. Participant taste consistency by preference ratio. The x-axis is a participant\u2019s relative M&amp;M preference ratio. The y-axis is the average of the standard deviation of their USA M&amp;M scores and the standard deviation of their Denmark M&amp;M scores. A value of 0 on the y-axis indicates perfect consistency in responses, while higher values indicate more inconsistent responses. Figure made with Altair.<\/sup><\/p>\n<p class=\"wp-block-paragraph\" id=\"4d7b\">Most participants were somewhat inconsistent in their ratings, ranking the same M&amp;M type differently across the 5 samples. This would be expected if the taste difference between European-sourced and American-sourced M&amp;Ms is not actually all that perceptible. Most inconsistent were participants who gave the same M&amp;M type \u201cGood\u201d, \u201cNormal\u201d,\u00a0<em>and<\/em>\u00a0\u201cBad\u201d responses (e.g., points high on the y-axis, with wider standard deviations of taste scores), indicating lower taste perception abilities.<\/p>\n<p class=\"wp-block-paragraph\" id=\"a6ed\">Intriguingly, four participants \u2014 one from each continent group \u2014 were perfectly consistent: they reported the same taste response for each of the 5 M&amp;Ms from each M&amp;M type, resulting in an average standard deviation of 0.0 (bottom of Figure 9). Excluding the one of the four who simply rated all 10 M&amp;Ms as \u201cNormal\u201d, the other three appeared to be \u201cPerfect Classifiers\u201d \u2014 either rating all M&amp;Ms of one type \u201cGood\u201d and the other \u201cNormal\u201d, or rating all M&amp;Ms of one type \u201cNormal\u201d and the other \u201cBad.\u201d Perhaps these folks are \u201csuper tasters.\u201d<\/p>\n<h3 class=\"wp-block-heading\" id=\"811c\">3.6 M&amp;M color<\/h3>\n<p class=\"wp-block-paragraph\" id=\"61a6\">Another possible explanation for the inconsistency in individual taste responses is that there exists a perceptible taste difference based on the M&amp;M color. Visually, the USA M&amp;Ms were noticeably more smooth and vibrant than the Denmark M&amp;Ms, which were somewhat more \u201csplotchy\u201d in appearance (Figure 10A). M&amp;M color was recorded during the experiment, and although balanced sampling was not formally built into the experimental design, colors seemed to be sampled roughly evenly, with the exception of Blue USA M&amp;Ms, which were oversampled (Figure 10B).<\/p>\n<figure class=\"wp-block-image size-large\"><img data-recalc-dims=\"1\" data-dominant-color=\"c3bcae\" data-has-transparency=\"false\" style=\"--dominant-color: #c3bcae;\" loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"827\" src=\"https:\/\/i0.wp.com\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/1_1MHZGfeuN3vCHrh20xb7rw-1024x827.webp?resize=1024%2C827&#038;ssl=1\" alt=\"\" class=\"wp-image-598335 not-transparent\" srcset=\"https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/1_1MHZGfeuN3vCHrh20xb7rw-1024x827.webp 1024w, https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/1_1MHZGfeuN3vCHrh20xb7rw-300x242.webp 300w, https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/1_1MHZGfeuN3vCHrh20xb7rw-768x620.webp 768w, https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/1_1MHZGfeuN3vCHrh20xb7rw.webp 1400w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\"><figcaption class=\"wp-element-caption\">Figure 10. M&amp;M colors.\u00a0<strong>A.\u00a0<\/strong>Photo of each M&amp;M color of each type. It\u2019s perhaps a bit hard to perceive on screen in my unprofessionally lit photo, but with the naked eye, USA M&amp;Ms seemed to be brighter and more uniformly colored while Denmark M&amp;Ms have a duller and more mottled color. Is it just me, or can you already hear the Europeans saying \u201cThey are brighter because of all those extra chemicals you put in your food that we ban here!\u201d\u00a0<strong>B.\u00a0<\/strong>Distribution of M&amp;Ms of each color sampled over the course of the experiment. The Blue USA M&amp;Ms were not intentionally oversampled \u2014 they must be especially bright\/tempting to experimenters. Figure made with Altair.<\/figcaption><\/figure>\n<p class=\"wp-block-paragraph\" id=\"29e8\">We briefly visualized possible differences in taste responses based on color (Figure 11), however we do not believe there are enough data to support firm conclusions. After all, on average each participant would likely only taste 5 of the 6 M&amp;M colors once, and 1 color not at all. We leave further M&amp;M color investigations to future work.<\/p>\n<figure class=\"wp-block-image size-large\"><img data-recalc-dims=\"1\" data-dominant-color=\"dbe3d8\" data-has-transparency=\"false\" style=\"--dominant-color: #dbe3d8;\" loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"783\" src=\"https:\/\/i0.wp.com\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/0_jp4GM2NUu0QwaL8q-1-1024x783.webp?resize=1024%2C783&#038;ssl=1\" alt=\"\" class=\"wp-image-598337 not-transparent\" srcset=\"https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/0_jp4GM2NUu0QwaL8q-1-1024x783.webp 1024w, https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/0_jp4GM2NUu0QwaL8q-1-300x230.webp 300w, https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/0_jp4GM2NUu0QwaL8q-1-768x588.webp 768w, https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/0_jp4GM2NUu0QwaL8q-1.webp 1400w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\"><figcaption class=\"wp-element-caption\">Figure 11. Taste response profiles for M&amp;Ms of each color and type. Profiles are reported as percentages of \u201cBad\u201d, \u201cNormal\u201d, and \u201cGood\u201d responses, though not all M&amp;Ms were sampled exactly evenly. Figure made with Altair.<\/figcaption><\/figure>\n<h3 class=\"wp-block-heading\" id=\"a4ac\">3.7 Colorful commentary<\/h3>\n<p class=\"wp-block-paragraph\" id=\"dadc\">We assured each participant that there was no \u201cright \u201canswer\u201d in this experiment and that all feelings are valid. While some participants took this to heart and occasionally spent over a minute deeply savoring each M&amp;M and evaluating it as if they were a sommelier, many participants seemed to view the experiment as a competition (which occasionally led to deflated or inflated pride). Experimenters wrote down quotes and notes in conjunction with M&amp;M responses, some of which were a bit \u201ccolorful.\u201d We provide a hastily rendered word cloud for each M&amp;M type for entertainment purposes (Figure 12) though we caution against reading too far into them without diligent sentiment analysis.<\/p>\n<figure class=\"wp-block-image size-large\"><img data-recalc-dims=\"1\" data-dominant-color=\"d5e8e4\" data-has-transparency=\"false\" style=\"--dominant-color: #d5e8e4;\" loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"312\" src=\"https:\/\/i0.wp.com\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/1_UlSrCAKuoCKdAJSuNAUaHw-1024x312.webp?resize=1024%2C312&#038;ssl=1\" alt=\"\" class=\"wp-image-598338 not-transparent\" srcset=\"https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/1_UlSrCAKuoCKdAJSuNAUaHw-1024x312.webp 1024w, https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/1_UlSrCAKuoCKdAJSuNAUaHw-300x92.webp 300w, https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/1_UlSrCAKuoCKdAJSuNAUaHw-768x234.webp 768w, https:\/\/towardsdatascience.com\/wp-content\/uploads\/2025\/02\/1_UlSrCAKuoCKdAJSuNAUaHw.webp 1400w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\"><figcaption class=\"wp-element-caption\">Figure 11. A simple word cloud generated from the notes column of each M&amp;M type. Fair warning \u2014 these have not been properly analyzed for sentiment and some inappropriate language was recorded. Figure made with WordCloud.<\/figcaption><\/figure>\n<h2 class=\"wp-block-heading\" id=\"f494\">4. Conclusion<\/h2>\n<p class=\"wp-block-paragraph\" id=\"a923\">Overall, there does not appear to be a \u201cglobal consensus\u201d that European M&amp;Ms are better than American M&amp;Ms. However, European participants tended to more strongly express negative reactions to USA M&amp;Ms while North American participants seemed relatively split on whether they preferred M&amp;Ms sourced from the USA vs from Europe. The preference trends of Asian participants often fell somewhere between the North Americans and Europeans.<\/p>\n<p class=\"wp-block-paragraph\" id=\"cf1a\">Therefore, I\u2019ll admit that it\u2019s probable that Europeans are not engaged in a grand coordinated lie about M&amp;Ms. The skew of most European participants towards Denmark M&amp;Ms is compelling, especially since I was the experimenter who personally collected much of the taste response data. If they found a way to cheat, it was done well enough to exceed my own passive perception such that I didn\u2019t notice. However, based on this study, it would appear that a strongly negative \u201cvomit flavor\u201d is not universally perceived and does not become apparent to non-Europeans when tasting both M&amp;Ms types side by side.<\/p>\n<p class=\"wp-block-paragraph\" id=\"496e\">We hope this study has been illuminating! We would look forward to extensions of this work with improved participant sampling, additional M&amp;M types sourced from other continents, and deeper investigations into possible taste differences due to color.<\/p>\n<p class=\"wp-block-paragraph\" id=\"7964\">Thank you to everyone who participated and ate M&amp;Ms in the name of science!<\/p>\n<p class=\"wp-block-paragraph\" id=\"5747\">Figures and analysis can be found on github:\u00a0<a href=\"https:\/\/github.com\/erinhwilson\/mnm-taste-test\" rel=\"noreferrer noopener\" target=\"_blank\">https:\/\/github.com\/erinhwilson\/mnm-taste-test<\/a><\/p>\n<p class=\"wp-block-paragraph\" id=\"4cab\"><em>Article by Erin H. Wilson, Ph.D.[1,2,3] who decided the time between defending her dissertation and starting her next job would be best spent on this highly valuable analysis. Hopefully it is clear that this article is intended to be comedic\u2014 I do not actually harbor any negative feelings towards Europeans who don\u2019t like American M&amp;Ms, but enjoyed the chance to be sassy and poke fun at our lively debates with overly-enthusiastic data analysis.<\/em><\/p>\n<p class=\"wp-block-paragraph\" id=\"3c53\"><em>Shout out to Matt, Galen, Ameya, and Gian-Marco for assisting in data collection!<\/em><\/p>\n<p class=\"wp-block-paragraph\" id=\"3075\"><em>[1] Former Ph.D. student in the Paul G. Allen School of Computer Science and Engineering at the University of Washington<\/em><\/p>\n<p class=\"wp-block-paragraph\" id=\"302c\"><em>[2] Former visiting Ph.D. student at the Novo Nordisk Foundation Center for Biosustainability at the Technical University of Denmark<\/em><\/p>\n<p class=\"wp-block-paragraph\" id=\"7735\"><em>[3] Future data scientist at LanzaTech<\/em><\/p>\n<p>The post <a href=\"https:\/\/towardsdatascience.com\/do-european-mms-actually-taste-better-than-american-mms\/\">Do European M&amp;Ms Actually Taste Better than American M&amp;Ms?<\/a> appeared first on <a href=\"https:\/\/towardsdatascience.com\/\">Towards Data Science<\/a>.<\/p>\n<\/div>\n<p> \t<BR><br \/>\n <BR><\/BR><br \/>\n    Erin Wilson<br \/>\n \t<BR><br \/>\n<BR><\/BR><br \/>\n<a href=\"https:\/\/towardsdatascience.com\/do-european-mms-actually-taste-better-than-american-mms\/\">Go to original source<\/a><br \/>\n \t<BR><br \/>\n <BR><\/BR><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Do European M&#038;Ms Actually Taste Better than American M&#038;Ms? (Oh, I am the only one who\u2019s been asking this question\u2026? Hm. Well, if you have a minute, please enjoy this exploratory Data Analysis \u2014 featuring experimental design, statistics, and interactive visualization \u2014 applied a bit too earnestly to resolve an international debate.) 1. Introduction 1.1 [&hellip;]<\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[62,211,83,82,1827,1828,238],"tags":[1831,1830,1829],"class_list":["post-2002","post","type-post","status-publish","format-standard","hentry","category-aimldsaimlds","category-data-analysis","category-data-science","category-data-visualization","category-editors-picks","category-food-science","category-statistics","tag-american","tag-chocolate","tag-ms"],"_links":{"self":[{"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/posts\/2002"}],"collection":[{"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/comments?post=2002"}],"version-history":[{"count":0,"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/posts\/2002\/revisions"}],"wp:attachment":[{"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/media?parent=2002"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/categories?post=2002"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/tags?post=2002"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}