Collecting and processing data practice questions.

In a genetics experiment, why would a student use a chi-squared test on the counts of offspring phenotypes?

• To compare observed ratios with expected ratios
• To calculate the mean number of offspring
• To measure how spread out the offspring data are
• To prove that one allele is dominant

Answer: To compare observed ratios with expected ratios

A chi-squared test checks whether the difference between observed and expected category counts is greater than expected by chance; it does not measure means, spread, or prove dominance.

A scatter graph of two variables gives a correlation coefficient (r) of -0.9. How is this relationship best described?

• A strong negative correlation
• A strong positive correlation
• A weak negative correlation
• No correlation at all

Answer: A strong negative correlation

A value of r close to -1 shows a strong negative correlation: as one variable increases the other tends to decrease. The sign gives the direction and the size gives the strength.

In an osmosis investigation the results are processed as percentage change in mass rather than as change in grams. Why?

• The cylinders did not all start at the same mass
• Percentages are easier to plot on a graph than masses
• The balance is more accurate over a percentage range
• The change in grams would be too small to be measured

Answer: The cylinders did not all start at the same mass

A gain of 0.6 g means something quite different for a 5 g cylinder and a 20 g one, so absolute changes cannot be pooled or compared. Processing that removes an irrelevant difference between specimens is what makes replicates comparable in the first place.

A balance displays masses to two decimal places, but the student writes each reading down to the nearest whole gram. What is the consequence?

• Precision the instrument supplied has been thrown away
• The readings become systematically higher than the true values
• The mean of the readings can no longer be calculated
• The uncertainty of the balance has itself been increased

Answer: Precision the instrument supplied has been thrown away

Rounding at the point of recording destroys information that cannot be recovered later, and it can hide a real difference between treatments entirely. The balance is unaffected; what has changed is the record of what it said.

A student measures the same leaf ten times and obtains values between 47.9 and 48.1 mm. The leaf is in fact 52.0 mm long. How is this set of measurements described?

• Precise but not accurate
• Accurate but not precise
• Neither accurate nor precise
• Both accurate and precise

Answer: Precise but not accurate

The readings agree closely with one another, so they are precise, but they all sit four millimetres below the true value, so they are not accurate. That combination almost always points to a systematic fault such as a mis-set zero rather than to careless reading.

Why should the scales of a graph be chosen so that the plotted points fill most of the available grid?

• A compressed plot conceals both the trend and the scatter
• Points near the edge of the grid are read more accurately
• A full grid is required before a line of best fit may be drawn
• The gradient can only be calculated from a full-sized plot

Answer: A compressed plot conceals both the trend and the scatter

Squeezing the data into a corner makes a real curve look like a straight line and makes genuine scatter invisible. The choice of scale is a decision about what the reader will be able to see, which is why an axis that starts far below the lowest value is treated as a fault.