Professional Documents
Culture Documents
• You want to know how temperature affects soil respiration. Specifically, you ask how
increased air temperature near the soil surface affects the amount of carbon dioxide
(CO2) respired from the soil.
• How you manipulate the independent variable can affect the experiment’s external
validity – that is, the extent to which the results can be generalized and applied to the
broader world.
• Soil-warming experiment
• just slightly above the natural range for your study region.
• Second, you may need to choose how finely to vary your independent variable.
Sometimes this choice is made for you by your experimental system, but often you
will need to decide, and this will affect how much you can infer from your results.
• Phone-use experiment
• a categorical variable: either as binary (yes/no) or as levels of a factor (no phone use,
low phone use, high phone use).
• First, you need to consider the study size: how many individuals will be included in
the experiment? In general, the more subjects you include, the greater your
experiment’s statistical power, which determines how much confidence you can have
in your results.
• Then you need to randomly assign your subjects to treatment groups. Each group
receives a different level of the treatment (e.g. no phone use, low phone use, high
phone use).
• You should also include a control group, which receives no treatment. The control
group tells us what would have happened to your test subjects without any
experimental intervention.
• When assigning your subjects to groups, there are two main choices you need to
make:
• Randomization
• In a randomized block design (aka stratified random design), subjects are first
grouped according to a characteristic they share, and then randomly assigned to
treatments within those groups.
The 7 steps of experimental
design
• Terms in this set (6)
• State conclusion.
1. Make a hypothesis.
Let's say you want to make a marketing email that will improve
engagement rates. A good hypothesis for this might be, "Making an email
with emojis in both the subject line and copy will increase our engagement
rates by at least 25%." This is a good hypothesis because you can prove or
disprove it, it isn't subjective, and it has a clear measurement of
achievement.
2. Collect research.
After creating your hypothesis, begin to gather research. Doing this will give
you background knowledge about experiments that have already been
conducted and get an idea of possible outcomes.
Once you've collected the research, you can choose which avenue you will
take and what metrics to measure.
For instance, maybe you will run an A/B test. This method will allow you to
measure the results of two different emails, and figure out which email
performs better with your target audience..
Finally, choose your recipients and conduct the experiment. Next, you'll
analyze your results.
Once you've run the experiment, collect and analyze the results. Use the
metrics you've decided upon in the second step and conclude if your
hypothesis was correct or not.
The prime indicators for success will be the metrics you chose to focus on.
For instance, for the marketing email example, did engagement numbers
appear higher? If the CTR, impressions, and click-to-open rates are at or
higher than the 25% goal, the experiment would be considered one where
the hypothesis was accepted.
Now that you know how to conduct a marketing experiment, let's go over a
few different ways to run them.
Types of Marketing Experiments
There are many types of marketing experiments you can conduct with your
team. These tests will help you determine how aspects of your campaign
will perform before you roll out the campaign as a whole.
1. A/B Testing
HubSpot's email tool offers an A/B test feature for Professional and
Enterprise users. Alternatively, check out 8 of the Best A/B Testing Tools
for 2019 for options of other tools to help you perform A/B tests.
This method is useful because you can better understand the preferences
of users who will be using your product.
2. Different CTAs
Experimenting with different CTAs can improve the number of people who
engage with your content. For instance, instead of using "Buy now!" to pull
customers in, why not try, "Learn more?"
You can also test different colors of CTAs as opposed to copy. Another
CTA factor that I've been seeing around are ones that are animated.
To learn more about different types of CTAs, check out 8 Types of CTAs
You Should Absolutely Try on Your Blog.
For a CTA-related marketing experiment, you'll want to either use PPC ads
or landing pages to insert your CTA . From there, measure relevant metrics
based on your hypothesis and design of the button.
3. Animated Ads
As a big purveyor of GIFs in the workplace, animating ads are a great way
to catch the attention of potential customers. Animating ads don't
necessarily mean using GIFs -- you might also try small videos or ads with
multiple cards, which can catch the attention of web browsers.
This Instagram ad from Buffer, above, uses multimedia to make their post
pop. If you're testing out PPC advertising, try diversifying those ads to
capture the interest of more audiences. Additionally, you might run different
types of copy with your ads to see which language compels your audience
to click.
LINK VIDEO:
https://www.instagram.com/p/B2hLPCtnKP1/?utm_source=ig_web_button_share_sheet
Is there a social media site you're not using? For instance, lifestyle brands
might prioritize Twitter and Instagram, but implementing Pinterest opens
the door for an untapped audience.
You might consider testing which hashtags or visuals you use on certain
social media sites to see how well they perform. The more you use certain
social platforms, the more you can iterate based on what your audience is
engaging with the most.
5. Experiment Globally
If you post on an Eastern Time Zone (ET) schedule, run an experiment that
involves Pacific or Central Time Zones.
You might even use your social media analytics to determine which
countries or regions you should focus on -- for instance, my Twitter
Analytics, below, demonstrates where most of my audience resides. If,
alternatively, I saw most of my audience came from India, I might need to
alter my social strategy to ensure I catered to India's Time Zone, as well.
When experimenting with different time zones, consider making content
specific to the audience you're trying to reach. If you're trying to reach global
audiences, why not post something in a few different languages? Alternatively,
if you have international offices, you might spotlight different employees from
your offices all over the globe.
Ultimately, marketing experiments are a cost-effective way to get a picture
of how new content ideas will work in your next campaign, which is critical
for ensuring you continue to delight your audience. For more new content
ideas, check out our ultimate round-up here.
Key Takeaways
1. Discussion: For each of the following topics, list the pros and cons of a between-subjects and
within-subjects design and decide which would be better.
1. You want to test the relative effectiveness of two training programs for running a marathon.
2. Using photographs of people as stimuli, you want to see if smiling people are perceived as more
intelligent than people who are not smiling.
3. In a field experiment, you want to see if the way a panhandler is dressed (neatly vs. sloppily)
affects whether or not passersby give him any money.
4. You want to see if concrete nouns (e.g., dog) are recalled better than abstract nouns (e.g., truth).
2. Discussion: Imagine that an experiment shows that participants who receive psychodynamic therapy
for a dog phobia improve more than participants in a no-treatment control group. Explain a
fundamental problem with this research design and at least two ways that it might be corrected.