Observation in the natural sciences means gathering information directly from a source. It includes noticing events or things and collecting data by directly interacting with what is being studied.

In living things, observation usually happens through the senses, like sight or hearing. In science, observation often uses tools to see, measure, and record information that humans cannot detect with their senses alone. This helps scientists study things that are too small, far away, or invisible to the eye.

In science, observations are usually divided into two types: qualitative and quantitative. Qualitative observations describe qualities, such as color or texture, while quantitative observations use numbers, like measurements or counts.

The word "observation" can describe both the act of looking and the information that is written down as a result of that act.

Science

The scientific method uses observations of natural events to create and test ideas. This process includes repeated steps to help scientists learn and improve their understanding:

• Ask a question about something happening in nature
• Observe the event carefully
• Create a guess that might answer the question
• Predict what would happen if the guess is correct
• Test the guess by doing experiments, watching closely, studying in the field, or using models
• Use collected information to decide if the guess is correct, change it, or make a new guess, then repeat the steps
• Write a clear description of how the observation was made and the results found
• Share the findings with other scientists who study the same topic

Each step depends on observations that others can repeat and check, which helps scientists make sure their results are correct.

Observations are used in step 2 and step 5. However, for results to be reliable, different people must see the same thing in the same way. Human senses can be unclear and give descriptions that are hard to compare. To solve this, scientists use measurements, which provide clear numbers that are easier to record and compare.

Measurement compares what is observed to a standard unit, such as a tool, a rule, or an agreement. This standard must be the same for everyone. The result of measurement is a number showing how many standard units match the observation.

Using numbers helps scientists write down observations clearly and compare them. If two observations have the same number, they are considered the same within the accuracy of the measurement.

Human senses have limits and can be wrong, like when seeing things that are not real. These limits make it hard to trust observations made without tools.

To help, scientists use tools like scales, clocks, telescopes, microscopes, thermometers, cameras, and recorders. These tools help measure things humans can see. Other tools, like dyes, voltmeters, spectrometers, infrared cameras, oscilloscopes, interferometers, Geiger counters, and radio receivers, help detect things humans cannot see.

A challenge in science is that observing something can change what is being studied. This is called the observer effect. For example, checking tire pressure by letting air out changes the pressure being measured.

Better tools can reduce how much observing changes the result. These tools help scientists study things with less interference.

All observations, whether done by people or tools, involve some kind of change. Observation is a process that cannot be reversed and increases disorder in the system being studied.

Paradoxes

In some areas of science, observations can differ based on factors that are usually not important in daily life. These differences are sometimes shown through "paradoxes," where an event seems different when viewed from two separate perspectives, even though it might seem to go against what is commonly expected.

• Relativity: In physics that studies very fast speeds, close to the speed of light, different people may see different results for things like length, time, and mass. This depends on how fast they are moving compared to the object being studied. For example, in the "twin paradox," one twin travels at a very high speed and returns younger than the twin who stayed on Earth. This happens because time moves more slowly for objects moving at very high speeds compared to those that are not. In this type of physics, all observations must be explained based on the movement and position of the observer.

• Quantum Mechanics: In physics that studies very small systems, like atoms and particles, it is impossible to observe a system without changing it. In this case, the person observing becomes part of the system being studied. These systems are described by something called a "wave function," which often includes a mix of different possible states. When an observation or measurement is made, the system is always found in one specific state, not a mix. The act of measuring causes the wave function to change, making the system settle into one definite state. This process is called observation or measurement, whether it is done intentionally in an experiment or not.

Biases

Human senses do not work like a fair recorder, such as a video camera. Perception happens through a complex, mostly unconscious process where some parts of what is seen or heard are chosen and remembered, while others are forgotten.

This process depends on a person's mental model of the world, called a schema in psychology. This model is shaped by past experiences. Information from the senses is understood and stored based on this model. When remembering, missing details may be filled in with information that matches the schema, a process called reconstructive memory.

How much attention a person pays to different parts of an experience depends on what they think is important. This can lead to two people seeing the same event but remembering it differently, possibly disagreeing on facts. This subjectivity is a known problem with eyewitness testimony, which studies have shown is often not reliable.

In science, careful methods are used to reduce these biases. These include writing detailed notes about experiments, clearly separating what is directly observed from what is guessed, and using techniques like blind or double-blind tests to avoid personal influence.

Some important ways human psychology can affect observations include:

• Main article: Streetlight effect
• Human observations often focus on confirming what a person expects to see, a tendency called confirmation bias. This can cause new discoveries to be missed, such as the discovery of X-rays. It can also lead to incorrect support for widely held beliefs, such as outdated ideas about race in the early 1900s.

Modern tools often process observations before showing results to humans. With computers handling more tasks, it can be hard to tell where observation ends and interpretation begins.

This is especially important in digital image processing, where images used in research may be edited to highlight certain details. While this can help show important information, it might also support a researcher’s guess, creating a type of bias that is hard to measure.

To address this, some journals have rules about how images can be edited in published research. Many systems save unprocessed data from sensors. Scientific best practices also require that original, unaltered images used in research be kept and shared when needed.