Wikipedia:Reference desk/Archives/Science/2018 July 5

Science desk
< July 4	<< Jun | July | Aug >>	July 6 >

Welcome to the Wikipedia Science Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

July 5

Scientifically, is it possible to improve IQ score?

Scientifically, is it possible to improve IQ score? Or what human have it's constant for him forever, no matter what he does to improve it? --2A02:ED0:6D6D:F300:7869:435F:5ED7:13CC (talk) 13:09, 5 July 2018 (UTC)

See Intelligence_quotient#Interventions. Ian.thomson (talk) 13:12, 5 July 2018 (UTC)

See Flynn effect. IQs are on the rise for society as a whole. Section 3 of the article gives several possible explanations for this increase. --Doroletho (talk) 14:44, 5 July 2018 (UTC)

If you look further you'll see there is some evidence the effect is ending or has ended and IQs are now decreasing if anything. Dmcq (talk) 11:21, 6 July 2018 (UTC)

In any event, Doroletho has conflated two questions here. The Flynn effect is about a general rise of IQs in large segments of contemporary populations, whereas the OP was asking if it is possible for an individual to improve their score. The answer there is yes and no. IQ tests remain tests even if they are more predicated in generalized problem solving than in the application of a particular knowledge base, so familiarity with the particular cognitive exercises and analytical problems any particular IQ test employs will improve facility with those tasks. So one who gains familiarity with a particular kind of IQ test can leverage marginally improved scores. However, this does not necessarily equate to an actual increase in IQ.

The answer to the OP's inquiry also varies a bit by particular test. IQ is a funny thing, because we don't really have a firm definition for it that is abstractly descriptive, only a statistical argument for its empirical existence. Modern cognitive science is divided as to validity of a strong g factor (that is, a a statistical observence of a correlation between aptitude between different cognitive tasks) but even its strongest proponents are unlikely to believe there is such a thing as strict "general intelligence"; we now recognize the multiplicity of mental tasks is so great, that any application of psychometrics is always going to be limited with regard to the scope of the conclusions which can be reached as to general intelligence. But yes, previous testing (and I will try to find the sourcing shortly) has demonstrated improvements to score on many historical IQ tests can be improved--up to a point where scores will plateau, of course. Snow ^{let's rap} 11:17, 8 July 2018 (UTC)

• My mother was a schoolteacher and administered several tests. She would also (in between crosswords and knitting) solve the papers for herself. So on an IQ test (or at least, the Cattell tests she was familiar with) she could regularly nail a 180+ score. Now she was smart, but not that smart. Andy Dingley (talk) 17:12, 5 July 2018 (UTC)

The article here on wiki is not clear in the light of these answers. --2A02:ED0:6D6D:F300:7869:435F:5ED7:13CC (talk) 00:42, 6 July 2018 (UTC)

Why not?? My mother did one of those societies tests and got 165 and one of my sons 183 so it certainly isn't unreasonable. Dmcq (talk) 11:30, 6 July 2018 (UTC)

Subjectively I'd say a person's cognitive ability can vary with their health, diet, mental outlook, and all sorts of other things. Exercise and mental stimulation count for a lot. 173.228.123.166 (talk) 08:23, 8 July 2018 (UTC)

Indeed. Also, for some live tests, the competency of the psychologist is a factor. Moreover, people tend to do much better at particular tests (and usually do better at the same test the second time). Although most tests attempt to properly distribute different problem types, they aren't exact science and the balance may not fit everyone. Assessing intelligence is not an easy problem... To answer the original poster: yes, one can do better with some research (into popular test types) and practice (playing related games). —PaleoNeonate – 11:42, 8 July 2018 (UTC)

Measles

This news article about a measles case in Portland says, "A person is considered immune to measles if they were born before 1957, had the measles previously, or have been fully vaccinated for measles." How does being born before 1957 afford immunity? Is this just a poorly worded sentence based on the assumption that most people born before 1957 are immune because they are likely to have gotten measles in their childhood? (In this case, the sentence would be more accurate if "were born before 1957" were simply removed from the sentence.) ANDREVV (talk) 13:20, 5 July 2018 (UTC)

It's because they will have lived through several epidemics and so are likely to be immune see [1] Richerman (talk) 14:28, 5 July 2018 (UTC)

The year is included because it is the date at which the (U.S.) Centers for Disease Control and Prevention advise that people do not need the measles vaccine booster - Nunh-huh 16:11, 5 July 2018 (UTC)

When 20 is not the double of 10 in a scale

How do you call a scale like Celsius, in which 20 degrees is not the double of 10? Or SPF, where SPF 30 does not protect 50% more than SPF 20? --Doroletho (talk) 14:36, 5 July 2018 (UTC)

Celsius is a relative temperature scale with arbitrary reference points, conventionally 0 °C and 100 °C for the freezing and boiling temperatures of water. One may however convert a Celsius reading to Kelvin using [K] = [°C] + 273.15. The Kelvin scale is absolute so "twice the Kelvin" means "twice as hot".

SPF (Sun Protection Factor) is a reciprocal expression of the fraction of sunburn-producing UV rays that a Sunscreen allows to reach the skin. For example, "SPF 15" means that ​1⁄15 of the radiation reaches the skin, or conversely that one may sunbathe 15 times longer with the same risk of burn as without the sunscreen. SPF 30 should allow 50% longer sunbathing than SPF 20. DroneB (talk) 16:55, 5 July 2018 (UTC)

Our article Level of measurement explains the relevant concepts. It distinguishes four types of scale: nominal, ordinal, interval, and ratio. Celsius is an interval scale. SPF is a ratio scale. Looie496 (talk) 17:15, 5 July 2018 (UTC)

So in the terminology of that article, the answer to the question is "it's not a ratio scale". However, note that SPF is a ratio scale: SPF 30 does protect 50% more than SPF 20, if everything else is equal.

However, while I don't imagine that the authors of the article made it up, I do not think this "ratio scale" terminology is widely used. I read quite a bit about units of measurement and I don't remember ever hearing of it before. --76.69.47.228 (talk) 02:09, 6 July 2018 (UTC)

IP is correct. See Level of measurement. The terms you are likely looking for are nominal, ordinal, interval, and ratio scales. Nominal scales are things like gender, religious affiliation, or race. Ordinal scales are ones that can be ordered from most to least or largest to smallest, like likert scales, but the distance between each value is unknown. Interval scales are rather uncommon but examples are Fahrenheit and Celsius and dress sizes. They are like ordinal scales butt the distance between each value is known to be equal. However, they do not have a true zero value. Ratio scales have all the attributes of interval scales but have a true zero value. An example would be the Kelvin scale. EvergreenFir (talk) 02:15, 6 July 2018 (UTC)

That's my take too; even in the Kelvin scale you can't say "twice as hot". There'd have to be a scale based on the average velocities of the particles. Abductive (reasoning) 23:07, 6 July 2018 (UTC)

As an aside, I've been following this thread and find it very intriguing/educational. Thanks to all who've participated, even though I'm not the OP! Killiondude (talk) 23:21, 6 July 2018 (UTC)

I've heard of million, billion, trillion, etc. How many is a killion? ←Baseball Bugs ^{What's up, Doc?} carrots→ 01:52, 7 July 2018 (UTC)

Unless I'm misunderstanding the question, any logarithmic scale fits the description. Most commonly used ones are base-10 logarithmic, meaning each level on the scale represents a power of ten. A pH of 5 is ten times more acidic than a pH of 6, and so forth. --47.146.63.87 (talk) 06:41, 7 July 2018 (UTC)

I would use this mathematical terminology: 20 degrees celsius is not "double" the temperature of 10 degrees celsius, because the Celsius scale has a bias. In this sense, a bias is an offset added to a linear equation; bias is the "b" term in the common notation "y = mx + b. In the case of Celsius, that bias yields the difference between the Celsius scale and the Kelvin scale; its value is about 273.15; it is no coincidence that it relates to the value of absolute zero when measured in Celsius.

In particular, the linear equation here is the one that defines temperature (in celsius) as a function of the average kinetic energy in the molecules of a sample. It is a simple algebraic rearrangement of the definition of thermodynamic temperature, by solving for T, and biasing for the units in Celsius.

This is categorically different from some of the other explanations presented in the earlier discussion, because the scale is linear, not logarithmic or exponential or anything else. It is for this reason that a 10 degrees celsius difference is related to the same amount of change in heat energy: it takes the same amount of fuel ("ΔQ") to cook water from 10-to-20 °C as we require to cook it from 20-to-30 °C, or from 80-to-90 °C. These temperatures represent linear increases in the heat energy added to a sample; and, because of the scale's bias, zero celsius does not mean zero energy is contained in the sample.

Nimur (talk) 15:18, 8 July 2018 (UTC)

Celsius temperature is an affine function of absolute temperature, not a linear one. 173.228.123.166 (talk) 15:22, 8 July 2018 (UTC)

This is true; I do not disagree. An equation can be expressed as a linear polynomial; this is a different mathematical statement than saying that the system is linear. Quoting our article:

"Note that this usage of the term linear is not the same as in the section above, because linear polynomials over the real numbers do not in general satisfy either additivity or homogeneity. In fact, they do so if and only if b = 0. Hence, if b ≠ 0, the function is often called an affine function (see in greater generality affine transformation)."

For this reason - to expound on the subtle details! - mathematical physicists usually study linear systems for at least a half a year during their formal education. Here was the book I used: Haykin & Van Veen, Signals and Systems (2003). The first two hundred pages are on various definitions, and properties, of linear equations and linear systems; the most important properties are superposition and homogeneity.

We might say that energy and temperature satisfy a homogeneous linear relation if we measure temperature in Kelvins, but not so if we measure in Celsius.

The original question was about terminology used to describe different scales of measurement; suffice to say, the more terms you know, the more precisely you can describe the details - at the risk of losing the audience in domain-specific jargon that sounds an awful lot like technobabble-nonsense!

Nimur (talk) 15:31, 8 July 2018 (UTC)

Children of homosexual couples.

I recently came across an entry for Linda Perry. In her personal section it says her wife gave birth to their child. This seems to imply that biologically something happened. I do realize that this is a sensitive topic, however, for accuracy wouldn't it be more appropriate to specify surrogacy (or another method?) There may come a time where science allows for a biological child, but for now it is technically incorrect to state or imply that this child belongs biologically to both partners

https://en.wikipedia.org/wiki/Linda_Perry

https://en.wikipedia.org/wiki/Sara_Gilbert — Preceding unsigned comment added by 24.144.1.3 (talk) 17:31, 5 July 2018 (UTC)

This comment belongs on the article’s talk page, not here where the purpose is to request references. Loraof (talk) 19:27, 5 July 2018 (UTC)

This is addressed at Talk:Sara Gilbert#Parentage. -- ToE 07:53, 6 July 2018 (UTC)

Just to use the opportunity to educate the IP as to our policies though, I will note that the comments in that stale discussion are pretty off-base in terms of WP:Verification and WP:NPOV. The reason that it is permissible to mention the child as that of the couple is only indirectly connected to the legal, social, and common sense reasoning expressed on that talk page. As solid as some of those arguments are in various regards, they represent WP:original research just as much as the edit summaries of the IP who tried to remove that description in the first place. We don't use our own rational conclusions (however certain we are about them) to define how we describe people (or any topic) in our content, in Wikipedia's voice. The reason we need to describe the child as that of the couple is because that is how the sources describe the child and the relationship to his mothers (plural). For more information, OP, please see WP:WEIGHT. Snow ^{let's rap} 11:26, 8 July 2018 (UTC)