Latest / The Psychology Undergrad Podcast / Operational Definitions & Levels of Measurement: How Psychology Measures Constructs
Transcript
- 0:00Welcome to the Psychology undergrad. We are so
- 0:02excited to dive into our notes today. Yeah, we
- 0:05really are. So I was thinking, if you're cooking
- 0:08like a Thanksgiving turkey, knowing when it's
- 0:10done is actually beautifully straightforward.
- 0:13Right. You don't really have to guess. Exactly.
- 0:14You just jab a meat thermometer into it, boom,
- 0:17165 degrees, and it's safe to eat. You don't
- 0:20have to, like, debate the turkey's feelings or
- 0:23interpret its behavior. No, you definitely don't.
- 0:26You just measure it. Yeah, or, you know, Consider
- 0:29a physics lab. If you need to calculate an object's
- 0:33potential energy in Earth's gravitational field,
- 0:36the universe gives you this incredibly clean
- 0:38formula. Oh, yeah, the classic mass times height
- 0:41thing. Exactly. You find its mass, you find its
- 0:43height, and you just multiply them together along
- 0:45with the gravitational acceleration of Earth,
- 0:47which is 9 .8 meters per second squared, right?
- 0:51Yeah, a constant 9 .8. So you plug the numbers
- 0:54in and you get a totally objective score. Right.
- 0:56But how on earth do you measure human nature?
- 0:59I mean, you can't stick a thermometer into someone's
- 1:01ear to see how extroverted they are. That would
- 1:04make our field a lot easier, though. Right. Or,
- 1:06you know, you can't put a measuring tape around
- 1:09someone's skull to find out if they're depressed
- 1:11or if they hold a specific attitude toward taxes.
- 1:14No, you really can't. So to you, our third roommate
- 1:17listening in on this study session, that is our
- 1:21mission today. We are... unpacking how psychologists
- 1:24take the invisible, messy reality of human behavior
- 1:28and somehow assign a rigid mathematical number
- 1:32to it. It's really the ultimate challenge of
- 1:35psychometrics, because when we talk about psychological
- 1:37measurement, we aren't usually talking about
- 1:39physical rulers. OK, let's unpack this. What
- 1:42exactly are we talking about, then? Well, at
- 1:43its most basic, measurement in this context is
- 1:46simply the systematic procedure of assigning
- 1:49scores to individuals. Okay, systematic being
- 1:52the key word there. Right, you assign scores
- 1:54so that they represent a characteristic of those
- 1:56individuals. But yeah, the absolute key word
- 1:58there is systematic. Got it. Which means we first
- 2:01have to separate the things we can easily measure
- 2:03from the things we, well, can't. When I was looking
- 2:06through a textbook for this, it really helped
- 2:08me to divide human variables into two camps.
- 2:11OK, what's the first camp? So the first camp
- 2:13is simple variables. You know, sex, age, height,
- 2:17weight, birth order. The physical stuff. Yeah,
- 2:20like if I want to know your weight, I just point
- 2:23you to a bathroom scale. If I want your age,
- 2:25I check your birth certificate. Yeah, those are
- 2:27straightforward because they are directly observable.
- 2:30Exactly. But the vast majority of what makes
- 2:32psychology a unique science lives in the second
- 2:35camp, which we call constructs. Constructs. Yeah.
- 2:39Constructs are variables that cannot be observed
- 2:41directly because they are internal processes
- 2:43or behavioral tendencies. Like what? Like intelligence,
- 2:48self -esteem, fear. introversion. These are all
- 2:52constructs. Okay, a great way to visualize this
- 2:54for our notes, I think, is the most famous framework
- 2:57for organizing personality constructs. The big
- 3:01five. Oh, right. From Costa and McCray's work
- 3:03in 1992. Yep, 1992. You've probably heard the
- 3:07acronym OCEAN or CANOE sometimes. Right. Openness
- 3:11to experience, conscientiousness, extroversion,
- 3:14agreeableness, and neuroticism. Exactly. And
- 3:18Costa and McCray then broke each of those five
- 3:21massive dimensions down into six highly specific
- 3:25subcategories. Which they call facets. Yeah,
- 3:27facets. And measuring these facets is incredibly
- 3:30slippery. It's not like measuring a physical
- 3:32object where its properties are just, you know,
- 3:35constant. Definitely not. Thinking about this,
- 3:38the process actually mirrors how humanity has
- 3:41historically thought about measuring morality
- 3:43or religious devotion. Oh, that's an interesting
- 3:45way to look at it. Right. Because you can't put
- 3:47a human soul on a bathroom scale. You can't use
- 3:50a dipstick to check the volume of someone's goodness.
- 3:53No, you can't. You can only look for outward
- 3:55signs, the physical manifestations of an invisible
- 3:58state. Yeah, that makes a lot of sense. And to
- 4:00help remember Costa McCray's big five facets,
- 4:03look at the ones that sound exactly like these
- 4:05moral virtues. Oh yeah, there are a few that
- 4:07cross over. Yeah. Altruism is a facet measured
- 4:11under agreeableness. Values is a facet measured
- 4:14under openness and dutifulness is measured under
- 4:17conscientiousness. What's fascinating here is
- 4:20the danger of trying to pin down a construct
- 4:23like altruism or fear or extroversion. You cannot
- 4:27reduce it to one single isolated act. Right,
- 4:30because it's a pattern. Exactly. Constructs are
- 4:33summaries of complex unfolding behaviors over
- 4:36time. Okay, give me an example. Well, imagine
- 4:38a university student who scores in the 99th percentile
- 4:42for extroversion. OK, super extroverted. Right.
- 4:45But if a researcher observes her at this exact
- 4:47moment, she might be sitting perfectly still
- 4:50in a silent library reading a textbook for three
- 4:52hours. Oh, I see. If you base your measurement
- 4:55purely on that one single observation, you would
- 4:58categorize her as highly introverted. Because
- 5:00extroversion is a tendency across a variety of
- 5:03situations. It's not a permanent unbreaking state
- 5:06of being. Exactly. You have to capture the pattern,
- 5:07not just the snapshot. Right. We see the exact
- 5:10same challenge with cognitive constructs, like
- 5:12working memory capacity. Which is basically an
- 5:15entirely invisible internal mental workbench.
- 5:17Right. You can't see it. So to measure it systematically,
- 5:21cognitive psychologists use tasks like the backward
- 5:24digit span. Oh, I hate that test. It's so hard.
- 5:27A researcher will read a list of digits, say,
- 5:30seven, two, nine, and ask the participant to
- 5:33repeat them in reverse order. So, nine, two,
- 5:36seven. Exactly. And if they succeed... The list
- 5:39gets longer. The researcher keeps increasing
- 5:42the load until the participant's mental workbench
- 5:44just collapses and they make an error. And then
- 5:46the longest list they can successfully reverse
- 5:48becomes their score. Yep. The invisible construct
- 5:51is rendered visible through the failure point
- 5:54of the task. We see this in clinical psychology,
- 5:56too. Like if a clinician needs to measure a patient's
- 5:58depression, they obviously can't scan for a glowing
- 6:01red depression node in the brain. Unfortunately
- 6:03not. Right. So they might use the Beck depression
- 6:06inventory, which is a 21. Oh, yeah, the BDI.
- 6:10Yeah. The patient rates their loss of energy,
- 6:13feelings of sadness, or changes in sleep over
- 6:15the past two weeks. It's usually on a scale of
- 6:17zero to three. And then they just add those up.
- 6:19Exactly. The clinician sums those ratings up,
- 6:22giving a potential max score of 63. So that final
- 6:25sum is the systematic score representing the
- 6:28construct. All of this raises a massive practical
- 6:31hurdle for researchers, though. We know we're
- 6:34chasing these invisible ghosts, but to maintain
- 6:37scientific rigor, we have to define exactly what
- 6:40it is we are looking for before we start handing
- 6:43out questionnaires or testing memory. Hang on,
- 6:45I actually struggled with this concept in lecture.
- 6:48Oh really? Which part? Well, if researchers just
- 6:51need to define a word like neuroticism or intelligence,
- 6:55why can't they just open the Oxford English Dictionary?
- 6:57Oh, I see. Like, why do scientists need to invent
- 7:00an entirely separate, elaborate set of definitions
- 7:03just for their studies? Because Everyday language
- 7:06is just too informal for empirical science. Too
- 7:08informal. Yeah. Dictionaries reflect how the
- 7:11general public casually uses words and casual
- 7:13usage is just full of contradictions. That's
- 7:16fair. Plus, the dictionary doesn't even have
- 7:18terms for highly specific scientific mechanisms
- 7:21like working memory capacity. Right. You won't
- 7:24find that in Merriam -Webster. Exactly. Science
- 7:26requires concepts that can be empirically tested.
- 7:30debated and refined over time based on new data.
- 7:35So researchers separate their definitions into
- 7:36two phases, right? Yes. Phase one is the conceptual
- 7:39definition. Which is basically the underlying
- 7:41theory. Right. It describes the specific behaviors
- 7:44and internal processes that make up the construct.
- 7:47So conceptually, neuroticism isn't just being
- 7:51nervous, the way a dictionary might say. No,
- 7:53it's way more specific. Right. The scientific
- 7:56conceptual definition is the tendency to experience
- 7:58negative emotions like anxiety, anger, and sadness
- 8:02across a variety of situations. With the added
- 8:06theoretical understanding that it has a strong
- 8:08genetic component. Yes. And that it remains fairly
- 8:11stable throughout an individual's lifetime. So
- 8:13that's the conceptual definition. Phase two is
- 8:15where the rubber meets the road. The operational
- 8:17definition. Exactly. This specifies precisely
- 8:20how that conceptual variable is going to be measured
- 8:22in a given study. translating the abstract concept
- 8:26into a physical operation. And in psychology,
- 8:30these operations generally fall into three broad
- 8:32categories. Yeah, three main ones. Okay, category
- 8:35one is self -report measures. Which is essentially
- 8:37just asking participants to report on their own
- 8:40internal states. Right. The Rosenberg self -esteem
- 8:42scale is the classic benchmark here. It's a series
- 8:46of statements like, I feel that I have a number
- 8:49of good qualities, and the participant rates
- 8:51how strongly they agree or disagree. Yeah. And
- 8:54then category two moves away from asking and
- 8:56towards watching. So behavioral measures. OK.
- 8:59Observing them. Right. Researchers observe and
- 9:01record some aspect of a participant's behavior.
- 9:05The backward digit span task we mentioned is
- 9:08a highly structured laboratory behavioral measure.
- 9:11But if we want the ultimate Hall of Fame example
- 9:13of a behavioral measure, we have to talk about
- 9:15Albert Bandura's 1961 Bobo doll experiment. Oh
- 9:19yeah, an absolute classic. Right, so they were
- 9:21studying how children learn aggression. But aggression,
- 9:23conceptually, is a massive, messy idea. Huge.
- 9:27So Bandura's team created a brilliantly strict
- 9:29operational definition. They let kids play in
- 9:32a room containing a bobo doll. Which is like
- 9:35one of those inflatable clowns with a weighted
- 9:36base that pops back up when you knock it out.
- 9:38Yeah, exactly. And crucially, the researchers
- 9:40observing behind the two -way mirror weren't
- 9:42just writing down subjective notes like, oh,
- 9:44Timmy seems very angry today. Because that would
- 9:47be terrible science. Right. They operationally
- 9:50defined physical aggression by tallying highly
- 9:52specific predefined physical acts. Hitting the
- 9:56doll with a wooden mallet, punching it in the
- 9:59nose, kicking it across the room. Wow. Very specific.
- 10:04Yeah, the total count of those specific violence
- 10:07strikes during a 20 -minute period was the operational
- 10:10definition of aggression. It totally removes
- 10:12the observer's subjective bias. You don't have
- 10:15to guess the child's internal anger level, you
- 10:17just count the malice strikes. Exactly. Finally,
- 10:20we have category three, which is physiological
- 10:22measures. So, biology. Yeah. This involves recording
- 10:25the body's autonomous processes to infer psychological
- 10:28states. Researchers might track heart rate, blood
- 10:31pressure, conductance of the skin to measure
- 10:34sweating. Or hormone levels in saliva, right?
- 10:37Yep. Or even use fMRI machines to track blood
- 10:40flow to specific regions of the brain. You know,
- 10:43looking at these three methods, it loops right
- 10:45back to that idea of moral judgment we talked
- 10:48about. Oh, so. Well, whether it's counting a
- 10:50child's violent outbursts in a lab or measuring
- 10:54the cortisol spiking in someone's blood... You
- 10:56are judging the hidden internal state entirely
- 10:59by its physical fruits. Yeah, that's true. The
- 11:02problem with relying on those fruits, though,
- 11:05is that any single operational definition is
- 11:08inherently vulnerable. Every measurement tool
- 11:10has a blind spot. Right. Think about the physiological
- 11:13measure of heart rate. OK. If a researcher operationally
- 11:16defines my stress level solely by how fast my
- 11:19heart is beating, they're going to get skewed
- 11:22data if I just drank an entire pot of coffee.
- 11:24Oh, absolutely. My heart might be racing at 110
- 11:27beats per minute, but I'm not psychologically
- 11:30stressed or anxious. I'm just incredibly caffeinated.
- 11:33Right. The operational definition failed to capture
- 11:35the conceptual definition there. This is why
- 11:37modern psychological science rarely relies on
- 11:40just one method. To fix the flaws of individual
- 11:43measures, researchers use converging operations.
- 11:46Converging operations. Yeah. They deploy multiple
- 11:49operational definitions for the exact same construct
- 11:52within a study. or across a field of research.
- 11:56Let's use that concept of stress to see how this
- 11:59works in practice. Conceptually, stress is defined
- 12:02as an adaptive response to a perceived danger
- 12:05or threat involving a cascade of physiological,
- 12:08cognitive, and behavioral changes. Which is a
- 12:11massive footprint to cover. Huge. To capture
- 12:13that footprint, researchers will throw multiple
- 12:16distinct measurement tools at it. First, they
- 12:19might want to know your historical stress burden.
- 12:21Like what's happened to you recently. Right.
- 12:23They'd used the social readjustment rating scale,
- 12:26developed by Holmes and Rahi. This is a self
- 12:28-report checklist of major life events over the
- 12:30past year. But it's weighted, right? Exactly.
- 12:33Based on extensive previous research, going through
- 12:35a divorce requires a massive amount of psychological
- 12:38adaptation, so it's assigned 73 points. Changing
- 12:41jobs is 36 points. A major change in your sleep
- 12:45habits is 16 points. You sum up all the events
- 12:48a person has experienced to calculate their total
- 12:51life change units. But catastrophic life events
- 12:54don't capture the slow grinding stress of everyday
- 12:58life. I mean, someone might score a zero on that
- 13:00scale, but still be suffering. Right, which is
- 13:03a big blind spot. So the researchers will also
- 13:05administer the daily hassles and uplift scale.
- 13:08This captures the microstressors, you know, constantly
- 13:10misplacing your car keys, arguing with a roommate,
- 13:13or worrying about minor financial issues. Yeah.
- 13:15And then, recognizing that two people might experience
- 13:18the exact same life events but internalize them
- 13:21differently, they will add the perceived stress
- 13:24scale. Which measures what? it directly targets
- 13:26the internal cognitive state. So asking questions
- 13:29like, in the last month, how often have you felt
- 13:31nervous and stressed? And to anchor all of those
- 13:33self -reports and objective biology, they bring
- 13:36in the physiological measures. They take the
- 13:38participants' resting blood pressure. They swab
- 13:41their saliva to measure the concentration of
- 13:43the stress hormone cortisol. It's kind of like
- 13:45the mechanics of a smartphone's GPS. Oh, how
- 13:48so? Well, if your phone only connects to a single
- 13:51cell tower, The network only has a vague radius
- 13:54of where you might be. You could be anywhere
- 13:56within a three mile circle. Right. But if your
- 13:59phone connects to a second tower, the overlapping
- 14:01circles narrow down your location. When you bounce
- 14:04the signal off a third tower, the intersecting
- 14:07data pinpoints your location with incredible
- 14:10accuracy. That is the perfect way to remember
- 14:13converging operations. Researchers are bouncing
- 14:16their signals off a self -report tower, a behavioral
- 14:19tower, and a physiological tower to pinpoint
- 14:22the exact location of the psychological truth.
- 14:25Yeah, and we know this works beautifully in practice.
- 14:28If we connect this to the bigger picture, Suzanne
- 14:30Segerstrom and Gregory Miller published a massive
- 14:33meta -analysis in 2004. reading about that one?
- 14:36Yeah, they were looking at hundreds of studies
- 14:38on stress and the immune system. They found that
- 14:40whether researchers used the social readjustment
- 14:42scale, the daily hassle scale, or physiological
- 14:46cortisol measures. They all pointed to the same
- 14:48thing. Exactly. All of these converging operations
- 14:51correlated with a measurable drop in immune system
- 14:55functioning. Because all these different towers
- 14:57triangulated on the same reality, the scientific
- 15:00community can confidently state the broad conclusion.
- 15:03Chronic stress suppresses the immune system.
- 15:06It transforms a flimsy hypothesis into an undeniable
- 15:09scientific fact. Exactly. OK, so at this point
- 15:12in the study session, we've defined our invisible
- 15:14constructs. We've built precise operational definitions
- 15:18to drag those constructs into the physical world.
- 15:21And we've deployed converging operations to triangulate
- 15:24our data. Our lab notebooks are full of numbers.
- 15:27Right, full of numbers. But looking at all this
- 15:28data, I'm realizing a massive trap here. Oh.
- 15:32Yeah. If I rate my daily stress level a four
- 15:35on a questionnaire and you rate your daily stress
- 15:37an eight, my immediate instinct is to say you
- 15:40are exactly twice as stressed as I am. Oh, I
- 15:43see where you're going with this. But mathematically,
- 15:45considering how these numbers were generated,
- 15:47I don't think we are actually allowed to do that.
- 15:49You absolutely cannot do that. The way a researcher
- 15:52assigns scores completely dictates what they
- 15:55are mathematically permitted to do with those
- 15:57scores. Here's where it gets really interesting.
- 16:00Yeah, a number in a spreadsheet is not always
- 16:02just a number. It has properties. And this was
- 16:04S .S. Stevens' brilliant contribution to science
- 16:07in 1946. He realized that researchers were performing
- 16:10mathematical operations on data that didn't support
- 16:13those operations. Right. So he introduced the
- 16:16four levels of measurement. The acronym to memorize
- 16:19here is N .O .I .R. N .O .I .R. Nominal Ordinal
- 16:24Interval Ratio. Like Knorr Cinema. Exactly. And
- 16:28it operates like a staircase. Every time you
- 16:30step up to a higher level, the numbers hold more
- 16:33quantitative information, which unlocks more
- 16:35advanced statistical tools. Okay, let's start
- 16:37on the bottom step. N for nominal. The nominal
- 16:40level is strictly for categorical variables.
- 16:42The scores are purely category labels. There
- 16:44is zero magnitude and zero order. Right. It only
- 16:47tells you if two individuals belong to the same
- 16:50category or different categories. So if a researcher
- 16:52assigns the number one to participants whose
- 16:55favorite color is blue and the number two to
- 16:58participants whose favorite color is green, those
- 17:01numbers are totally arbitrary. Because green
- 17:04is not mathematically greater than blue. Exactly.
- 17:06Which means doing math on nominal data is absurd.
- 17:09You cannot add 1 and 2, divide by 2, and conclude
- 17:13that the average participant's favorite color
- 17:15is 1 .5. Right. 1 .5 colors doesn't make any
- 17:18sense. Right. The only statistical measure of
- 17:20central tendency you can use for nominal data
- 17:22is the mode. Which just means counting which
- 17:25category occurred most frequently. Exactly. So
- 17:28stepping up, we reach O for ordinal. The ordinal
- 17:31level assigns scores that represent a rank order.
- 17:34Now we finally know magnitude. We know if a participant
- 17:37is higher or lower on a specific variable. Right.
- 17:40Think of a marathon finish line. First place,
- 17:42second place, third place. We know the exact
- 17:45order of the runners. But there's a limitation,
- 17:47right? Yeah, the mathematical limitation of ordinal
- 17:49data is that the intervals between the ranks
- 17:51are not equal. The time gap between the first
- 17:54and second place runners might be a fraction
- 17:55of a second, but the second place runner might
- 17:58have beaten the third place runner by 10 full
- 18:00minutes. So the ranks don't tell us the actual
- 18:03distance between the scores. Exactly. We use
- 18:06ordinal data constantly in psychology though,
- 18:08especially in surveys. Imagine a customer satisfaction
- 18:12survey with the options. Very dissatisfied, somewhat
- 18:16dissatisfied, somewhat satisfied, very satisfied.
- 18:19Right, classic survey. Yeah, we can rank these.
- 18:22A very satisfied customer is happier than a somewhat
- 18:24satisfied customer. Obviously. But the psychological
- 18:27distance -like, the actual amount of attitude
- 18:30change between very dissatisfied and somewhat
- 18:33dissatisfied is not necessarily the exact same
- 18:36size as the gap between somewhat dissatisfied
- 18:39and somewhat satisfied. And because we don't
- 18:41know the exact distance between the ranks, our
- 18:44math is still limited. We can't confidently calculate
- 18:47a true average. Right. So we step up the staircase
- 18:50again to I for interval. The interval level finally
- 18:52gives us equal distances between the numbers.
- 18:55Equal intervals. Yes. The gap between a score
- 18:58of 10 and 20 is the exact same magnitude as the
- 19:01gap between 40 and 50. OK. That sounds pretty
- 19:04solid. It is B -U -T. And this is a mathematical
- 19:07trap I almost fell into. Earlier interval scales
- 19:10do not have a true zero. This raises an important
- 19:13question, though. What do you mean by a true
- 19:15zero? Well, the zero point on an interval scale
- 19:17is completely arbitrary. Oh, it doesn't represent
- 19:21the total absence of the construct. Exactly.
- 19:23The classic examples are the Fahrenheit and Celsius
- 19:26temperature scales. The physical difference in
- 19:29kinetic energy between 30 degrees and 40 degrees
- 19:32Fahrenheit is identical to the difference between
- 19:3480 and 90 degrees. So you have equal intervals.
- 19:37You can add and subtract them. Yes, but zero
- 19:39degrees Fahrenheit does not mean there is zero
- 19:41heat in the universe. It was just arbitrary baseline
- 19:44set by the scale's inventor. Oh, I see. This
- 19:47is why you cannot make ratio statements with
- 19:50interval data. You mathematically cannot say
- 19:52that an 80 -degree day is twice as hot as a 40
- 19:55-degree day. Since the zero point doesn't represent
- 19:57true emptiness... Multiplication and division
- 20:00are completely invalid. Exactly. In our field,
- 20:04IQ is generally treated as interval data. If
- 20:07someone scores a 140 on an IQ test and another
- 20:09person scores a 70, you cannot say the first
- 20:12person is twice as smart. Right, because an IQ
- 20:15of zero doesn't mean literally zero intelligence.
- 20:18The scale simply doesn't support that math. Exactly.
- 20:21Which brings us to the final step of the staircase,
- 20:24the most powerful level of measurement. for ratio
- 20:28ratio yes the ratio level has equal intervals
- 20:31but it also possesses a true absolute zero point
- 20:35that represents the complete absence of the quantity
- 20:37being measured. So like height in meters or weight
- 20:39in kilograms. Right. Or the Kelvin temperature
- 20:41scale, where zero actually means the absolute
- 20:44cessation of molecular movement. Wow. Literally
- 20:46zero. Yeah. Or, to make it even more grounded,
- 20:50the amount of money in your wallet, the number
- 20:51of siblings you have, or the total number of
- 20:53physical punches a child throws at a bobo doll
- 20:55in 20 minutes. Right. Because if you have zero
- 20:57siblings, it represents the absolute absence
- 21:00of siblings. Exactly. And because of that true
- 21:02zero, the ratio level unlocks all mathematical
- 21:05operations. You can finally make twice as much
- 21:08statements. So someone with six siblings has
- 21:10exactly twice as many as someone with three.
- 21:13Yep. A child who hits the doll ten times is twice
- 21:16as aggressive, operationally speaking, as a child
- 21:19who hits it five times. Understanding this in
- 21:22a wire hierarchy? completely changes how you
- 21:24view published research. I mean, the next time
- 21:27you see a headline claiming that people who do
- 21:29yoga are twice as happy as people who don't.
- 21:32Your immediate reaction shouldn't be to buy a
- 21:34yoga mat. No, your reaction should be to question
- 21:36the underlying statistics. Happiness is a psychological
- 21:39construct. How do they operationally define it?
- 21:42Right. If they used a self -report scale, which
- 21:44is ordinal or interval data at best, then claiming
- 21:48someone is twice as happy is a mathematical impossibility.
- 21:51It's basically a researcher treating interval
- 21:53data as if it were ratio data just to get a flashier
- 21:56headline. Exactly. So what does this all mean?
- 21:59Well, measurement isn't just a preliminary step
- 22:01before the real science begins. It is the very
- 22:03foundation that determines whether the subsequent
- 22:05conclusions are brilliant insights or, well,
- 22:09statistical fiction. We've covered incredible
- 22:11ground today. We started with invisible constructs,
- 22:14the ghosts of human nature, like extraversion,
- 22:17neuroticism, and altruism. We forced those concepts
- 22:21into the physical world by crafting rigorous
- 22:23operational definitions. We ensured accuracy
- 22:27by triangulating our methods using converging
- 22:29operations. And we protected our conclusions
- 22:31by strictly adhering to Stevens' four levels
- 22:34of measurement. But as we wrap up this study
- 22:37session, there's a lingering philosophical question
- 22:39that arises from NOIR. Oh. It comes back to the
- 22:43intersection of psychology and morality you mentioned
- 22:45earlier, doesn't it? Exactly. If we view human
- 22:48goodness or moral virtue as a psychological construct,
- 22:51we can operationally define it. We can measure
- 22:54how much money people donate or how many times
- 22:56they volunteer. We can rank them. But will we
- 22:58ever find a way to measure the human soul on
- 23:01a ratio scale? Oh, wow. Will we ever identify
- 23:05a true absolute zero of human goodness, or are
- 23:08our moral judgments always doomed to remain ordinal?
- 23:11Destined only to rank people against one another.
- 23:14Yeah, knowing who finished first or second, but
- 23:17never truly understanding the vast hidden psychological
- 23:20distance between us. Something to think about
- 23:22until our next session. Keep studying.