In Bash scripting, a general rule is to wrap variables and command substitution with double quotes to prevent strings from unintended splitting or globbing. However, it's not always correct. Consider the following script.

#!/bin/bash

for i in "$(seq 1 5)"
do
	echo "$i"
	echo "end"
done

exit 0

The for loop here is intended to act like for i in range(1,6) in Python. So, the expected output is

1
end
2
end
3
end
4
end
5
end

However, the actual output is

1
2
3
4
5
end

What goes wrong? We wrap $(seq 1 5) with double quotes to prevent it from unintended split and globbing. So, the Bash shell treats the output of the command as a single string, 1\n2\n3\n4\n5\n, resulting in only one iteration. In the correct script, the double quotes should be removed.

#!/bin/bash

for i in $(seq 1 5)
do
	echo "$i"
	echo "end"
done

exit 0

So, the double quotes always result in a monolithic element, right? Unfortunately, this is also not true. Consider the following script.

#!/bin/bash

arr=(1 2 3 4 5)
for i in "${arr[@]}"
do
	echo "$i"
	echo "end"
done

exit 0

The expect output is

1
2
3
4
5
end

However, the actual output is

1
end
2
end
3
end
4
end
5
end

This is the "at" expansion is little bit more special. "${arr[@]}" is expanded to "1" "2" "3" "4" "5". This explain the actual output. If you want to know more detail, consult man bash.

As a tangent, without double quotes, ${arr[@]} and ${arr[*]} expand to the same thing. However, with double quotes, "${arr[@]}" and "${arr[*]}" expand differently. The testing code and output are left as an exercise to the audience.