ACCURACY
PRECISION
Kenneth Vander C. Gervacio
CHARACTERISTICS
OF MEASUREMENTS
Accuracy vs Random and Significant Figures
Precision Systematic Errors
ACCURACY
Accurate is defined as the closeness of a value
to its true value, such as the closeness of an
arrow to the bull's eye at the center. Accurate
is correct, that is the bull's eye
How close a measurement is to the true or
accepted value.
“Correctness”
PRECISION
Precise is defined as the repeatability of the
values of a measurement, such as the closeness
of other arrows to the first one. Precise is
repeating, that is hitting the same spot, but that
may not even be the correct spot.
How consistent repeated measurements are,
regardless of how close they are to the true value.
“Consistency”
Accuracy and precision are
related because both describe
the quality of measurements,
but they focus on different
aspects of that quality. Both
deal with measurement
reliability. Together, they
provide trustworthy and
meaningful data
YOU CAN BE:
🎯Accurate & Precise
All darts are close to the bullseye and close
together
🎯 Precise, not Accurate
All darts are close together but far from the
bullseye
🎯 Accurate, not Precise
Darts are scattered but average close to the
bullseye
🎯 Neither Accurate nor Precise
Darts are scattered and far from the bullseye
ERRORS
Errors are the differences between a measured value and the true or
accepted value of a quantity.
No measurement is ever perfectly exact all the — all measurements
have some degree of uncertainty. Most often, even if one is a higly
skilled individual such as a pro archer or a pro basketball player, can’t
guarantee a consistent bullseye or hoop all the time even if they got
used to the distance from point A to point B. These are errors caused
by either the individual or the device used in a specific context.
TYPES OF
ERROR
1. RANDOM 2. SYSTEMATIC
ERRORS ERRORS
RANDOM ERRORS
a chance difference between the observed and true values of something
(e.g., a researcher misreading a weighing scale records an incorrect
measurement).
Random error is referred to as “noise”
Unpredictable variations in measurement results, usually small and caused
by chance or uncontrollable factors.
RANDOM ERRORS: LOW PRECISION
Some common sources of random error include:
natural variations in real world or experimental contexts.
individual differences between participants or units.
poorly controlled experimental procedures.
NATURAL VARIATIONS IN REAL WORLD OR
EXPERIMENTAL CONTEXTS
In an experiment about memory capacity, your participants are
scheduled for memory tests at different times of day. However,
some participants tend to perform better in the morning while
others perform better later in the day, so your measurements
do not reflect the true extent of memory capacity for each
individual.
INDIVIDUAL DIFFERENCES BETWEEN
PARTICIPANTS OR UNITS
You ask participants to administer a safe electric shock to
themselves and rate their pain level on a 7-point rating
scale. Because pain is subjective, it’s hard to reliably
measure. Some participants overstate their levels of pain,
while others understate their levels of pain.
REDUCING
RANDOM ERRORS
1. Take repeated 2. Increase your 3. Control
measurements sample size variables
A simple way to Large samples have In controlled
increase precision is less random errorthan small experiments, you
by taking repeated samples.That’s because the should carefully
measurements and errors in different directions control any
usingtheir average. cancel each other out more extraneous variables
efficiently when you have that could impact
more data points. your measurements
SYSTEMATIC ERROR
A consistent or proportional difference between the observed and true
values of something (e.g., a miscalibrated scale consistently registers
weights as higher than they actually are).
Systematic error is also referred to as BIAS.
These are consistent and predictable errors caused by faulty tools,
procedures, or bias.
They cause all results to be off in the same direction.
SYSTEMATIC ERRORS: LOW ACCURACY
Types of Systematic Errors:
Offset Error:
An offset error occurs when a scale isn’t calibrated to a correct zero point. It’s
also called an additive error or a zero-setting error.
Scale Factor Error:
A scale factor error occurs when measurements consistently differ from the
true value by a fixed proportion (e.g., always 10% too high or too low). This type
of error is also known as a correlational systematic error or a multiplier error, as
it affects all measurements in a predictable, proportional way.
SYSTEMATIC ERRORS: LOW ACCURACY
Types of Systematic Errors:
Offset Error Example:
Digital or Analog Weighing Scale:
At a grocery store, a customer places a bag of apples on the scale, which reads
3 kg, but the bag feels lighter. The clerk later discovers the scale wasn't reset
and had a 1 kg reading even when empty. This caused all items to appear 1 kg
heavier. Though the measurements were consistent, they were all inaccurate
by the same amount
SYSTEMATIC ERRORS: LOW ACCURACY
Types of Systematic Errors:
Scale Factor Example:
Digital or Analog Weighing Scale:
A weighing scale that consistently adds 10% to each weight introduces a scale
factor error, such that a true weight of 10 kg is recorded as 11 kg and a true
weight of 40 kg is recorded as 44 kg.
SYTEMATIC ERRORS: LOW ACCURACY
Some common sources of systematic error include:
Imperfections in measuring devices (e.g. poor calibrations, damaged
instruments, drift in electronic instruments over time, etc.)
Methodological Errors (e.g. improper setup, using an incorrect formula or
method of calculation, uncontrolled variables )
REDUCING
SYSTEMATIC ERRORS
1. Calibrate Measuring Instruments Regularly
Ensure instruments measure true values correctly
across the entire range by using certified standard
references (e.g., calibration weights, reference
thermometers).
Perform calibration checks at multiple points
RANDOM
ERRORS VS.
SYSTEMATIC
ERRORS
ACTIVITY:
Directions: Analyze and evaluate each scenario
as to what type of error is being implied. Briefly
explain why do you think so of your response.
SCENARIO 1:
During a school athletics event, a teacher uses a stopwatch to
time students in a 100-meter dash. She starts timing each runner
manually, but due to her reaction time, she consistently clicks the
stopwatch about half a second after the race begins. Even though
the timing is fairly consistent across all races, every recorded time
ends up being longer than the actual run. This delay introduces
what type of error? Specify and explain why.
SCENARIO 2:
At a fruit stand, the weighing scale is faulty and multiplies the
actual weight by 1.5. A 2 kg item is displayed as 3 kg, showing an
error of 1 kg, while a 4 kg item is displayed as 6 kg, with an error of
2 kg. The heavier the item, the larger the error. What type of error
is being referred to. Specify and explain why.
SCENARIO 3:
In a chemistry class, a student is tasked with measuring 50 mL of
water using a graduated cylinder. Each time she pours, the surface
of the water (meniscus) slightly curves, and her eye level isn’t
always perfectly aligned. Sometimes she reads from a slight angle
above, and other times from below the correct line. As a result, her
measurements vary: one trial shows 49.5 mL, another shows 50.3
mL, and another 52.5 mL. These small, unpredictable differences
are caused by what type of error? Justify your response.
SIGNIFICANT FIGURES
Significant figures are used to report a value,
measured or calculated, to the correct
number of decimal places or digits that will
reflect the precision of the value. Significant
figures show how many digits in a
measurement are trustworthy and
meaningful.
SIGNIFICANT FIGURES
AND ACCURACY
More significant figures usually
indicate higher accuracy, especially
if they reflect the true value. But too
many sig figs can give a false sense
of accuracy if the instrument or
method can’t support them.
EXAMPLE:
A ruler that only
measures to the nearest
millimeter should report
12.3 cm, not 12.345 cm,
because the extra digits
are not accurate or not
significant.
RATIONALIZATION:
A ruler that only measures to the nearest millimeter (0.1 cm) has a
limited precision. This means it can accurately report measurements
like 12.3 cm, but not more precise values like 12.345 cm.
Reporting 12.345 cm suggests a level of accuracy (to the thousandth
of a centimeter) that the ruler cannot actually provide, because it
doesn’t have the scale markings or resolution to measure that finely.
SIGNIFICANT FIGURES
AND PRECISION
Significant figures reflect how
finely a measurement is made —
more digits = more detail = more
precision.
However, precision alone doesn't
guarantee accuracy
EXAMPLE:
If a balance gives you 3.456 g,
3.457 g, and 3.455 g, that’s
precise (many sig figs), but it
may still be inaccurate if the
balance has a calibration error
(systematic error *low accuracy).
SIGNIFICANT FIGURE RULES
1. All nonzero digits are significant.
(Ex. the value 211.8 has four significant figures)
2. All zeros that are found between nonzero digits are significant.
(Ex. the number 20,007, with three 0s between the 2 and 7, has a total of five significant
figures)
3. Leading zeros (to the left of the first nonzero digit) are not significant.
(Ex. A value such as 0.0085, for example, has two significant figures because the 0s before the
8 are placeholders and are not significant.)
4. Trailing zeros for a whole number that ends with a decimal point are significant.
(Ex. For example, a value written as 320. shows the decimal point, which indicates that the 0 to
the right of the 2 was measured; therefore, the value has a total of three significant figures. If
the decimal point was not written, then 320 would have only two significant figures.)
5. Trailing zeros to the right of the decimal place are significant.
(Ex. A value such as 12.000 has a total of five significant figures, since the 0s after the decimal
place have been measured to be zeros, indicating they are as significant as any other nonzero
digit.)
ACTIVITY
Direction: Determine the number of significant figures of the
following items.
0.00450 105,007
58.600 0.000432
473.9 450
7.6000
THANK
YOU!
