Improving preservice chemistry teachers’ content knowledge through intervention activities

Figures & data

Table 1. Examples of potential bonfire analogy attributes.

	Analogy (Bonfire)		Target (attraction between nucleus and an electron)
	Feature	Relationship	Feature	Relationship
Shared attributes	The size of the bonfire/heat received	The larger of the bonfire greater the heat received by each person from it.	Nuclear charge/attraction between electron and nucleus	The larger the nuclear charge the greater the attraction between electron and nucleus.
	The distance between bonfire and person	The greater the distance the smaller the heat received. The heat received by each person at the same distance is independent of the number people at that distance.	Distance between electron and nucleus	The greater the distance between electron and nucleus the smaller the attraction between electron and nucleus. The attraction between electron and nucleus is independent of the number of electrons at the same distance from the nucleus.
	Screening of the fire (by people nearer)	The more people screening the fire the less heat received.	Shielding	The greater the number of inner electrons shielding the less the attraction between shielded electrons and the nucleus.
Unshared attributes	Heat from people	The bonfire does not receive equal heat from people.	Attraction between nucleus and electron	Equal mutual attraction between nucleus and electron.
	Heat from people	People do not radiate heat equally nor is it dependent on size.	Repulsion between electrons	Mutual and equal repulsion between each electron (as charge is the same).

Table 2. The TFCL electronic test statements which use the ‘conservation of force’ AC and how each relates to the potential impact of the interventions.

Statements as they appeared on the screen with prompt for ‘true or false’ response by electronic voting	Alternative conception that might be indicated by a ‘true’ or ‘false’ response	Science needed to select the correct response	Possible effect of intervention
			Review Coulomb’s Law	Bonfire
S3 The third ionisation energy for this atom is greater than the second as there are less electrons in the shell to share the attraction from the nucleus.	True: Attraction between nucleus and a particular electron varies according to how many electrons are present	The third IE is greater than the second IE because the outermost electron is nearer the nucleus and force of attraction is therefore higher.	Helps as charge on outermost electron and nucleus remain the same – so changes in ‘attraction’ must have other cause.	Possible effect if analogy is used in a way that allows for cloud contraction to be modelled by crowd placed in concentric rings moving towards fire as loss of people allows ring to be smaller.
S6 After the atom is ionised, it then requires more energy to remove a second electron because once the first electron is removed the remaining electrons receive an extra share of the attraction from the nucleus.	True: The nucleus charge can direct it attractiveness according to the number of electrons.	The positive charge on the nucleus is the same, but the remaining electrons are closer once an electron is removed therefore attraction is greater.	Helps as charge on outermost electron and nucleus remain the same – so changes in ‘attraction’ must have other cause.	Helps challenge the alternative idea, but as heat experienced would remain the same when a person left from near by
S9 The 11 protons in the nucleus give rise to a certain amount of attractive force that is available to be shared between the electrons.	True: The nucleus charge can direct it attractiveness according to the number of electrons.	The protons act together as a ‘positive entity’ which attracts each electron according to Coulombs’ Law. Removal of electrons does not change the magnitude of the charges of nucleus or an electron.	Helps as charge on outermost electron and nucleus remain the same – so changes in ‘attraction’ must have other cause.	Helps challenge the alternative idea, but as heat experienced would remain the same when a person left from nearby implying subsequent ionisation energies in the same shell should remain the same.

Note: Bold text shows when intervention may support rejection of AC or use of scientific ideas.

Table 3. The TFCL electronic test statements which do not use ACs to allow for discrimination and how each relates to the potential impact of the interventions.

Statements as they appeared on the screen with prompt for ‘true or false’ response by electronic voting	Alternative conception that might be indicated by a ‘true’ or ‘false’ response	Science needed to select the correct response	Possible effect of intervention
			Review Coulomb’s Law	Bonfire
S1 Energy is required to remove an electron from the atom.	None	Energy is required to ionise atoms as work has to be done to overcome the mutual attraction between the electron and nucleus.	Small as question is very straightforward – connects attractions to the energy needed to overcome these forces.	None – Analogy may clarify attraction between electron and nucleus but there is no direct link to energy.
S7 After the atom is ionised, it then requires more energy to remove a second electron because the second electron experiences less shielding.	None	Shielding decreases with the number of electrons present. ‘Shielding’ describes a reduction in the ‘effective’ attraction between nucleus and electron caused by electron–electron interactions.	Helps to recognise that the source of ‘shielding’ is active (due to electron–electron repulsion) rather than passive and blocking in nature	Helps with a passive blocking view of shielding as in the analogy, people in between the fire and others block the heat felt.
S10 After this atom is ionised, it then requires more energy to remove a second electron because the second electron is in a lower energy level.	None	Second electron is in a lower energy level and hence would require more energy to remove	No effect No direct linkage to energy levels here	If energy levels are related to position – might help (small effect – if any)

Note: Bold text shows when intervention may support rejection of AC or use of scientific ideas.

Table 4. Framework showing the relationship between confidence and accuracy of true/false responses and also the coding and ordinal scales used for analysis.

Knowledge indicated by the response	Level of confidence in response	Coded (For discrimination analysis)	Concept grasp as indicated from the response and confidence	Ordinal Scale for quantitative analysis
Correct	Completely confident	C4	Scientific concept/rejection of AC	7
	Fairly sure	C3		6
	Unsure	C2	Unsure use of Scientific concept/rejection of AC	5
	Very unsure	C1	Guessing	4
Incorrect	Very unsure	I1
	Unsure	I2	Unsure use of AC	3
	Fairly sure	I3	AC used	2
	Completely confident	I4		1

Table 5. Comparison of discrimination indices for non-AC statements with AC statements.

Response pair for Wilcoxon Signed Ranks Test	Discrimination index for S1, 7 and 10 compared with discrimination index for S3, 6 and 9 (before intervention activities)
Z Based on positive ranks.	−7.170
Effect size r	0.75 (large)
Asymp. Sig.	<0.000
Median values (change)	1.67/−1.00 (2.67)
Median response descriptions	Confident giving of correct responses/confident giving of incorrect responses.

Table 6. Showing significance and effect size and median scores (on an ordinal scale of 1–7) for S3, S9 & S9 for differences in pre- and post-test.

Response pairs for Wilcoxon Signed Ranks Test	S3pre–S3post	S6pre–S6post	S9pre–S9post
Z Based on positive ranks.	4.849	5.479	4.828
Effect size r	0.357	0.404	0.356
Asymp. Sig. (two-tailed)	<0.001	<0.001	<0.001
Median values (change) (ordinal scale 1–7)	4/6 (+2)	3/6(+3)	3/5 (+2)
Median response change equivalent to (see  Table 4 for ordinal scale values)	Guessing to fairly sure correct	Unsure incorrect to fairly sure correct	Unsure incorrect to unsure correct

Table 7. Changes in discrimination pre and post-intervention activities.

Response pairs for Wilcoxon Signed Ranks Test	Discrimination index for Statements using the CF AC (S3, S6 and S9) pre compared with post-intervention	Discrimination index for Statements not using common ACs (S1, S7 and S10) compared with post-intervention	Discrimination index for Statements not using common ACs (S1, S7 & S10) compared with those using the CF AC (S3, S6 and S9) post-intervention
Z Based on negative ranks.	−5.913	−2.072
Effect size r	0.62	0.22
Asymp. Sig. (two-tailed)	<0.001	Not statistically significant	Not significant
Median values (change)	−1.00/1.50 (+2.50)
Median response change equivalent to	Confident in incorrect to confident in correct.	Difference is not significant	Difference is not significant

Figure 1. Summary of the effect of the intervention activities on responses.

Table 8. Effects of interventions on the ideas used to explain ionisation energies.

Ideas used to explain ionisation energy before intervention	Interviewed PSCT(Pseudonym)	Effect of intervention activities on ideas used to explain ionisation energies
		After intervention	Use of Bonfire analogy in interviews
Uses scientific causality in interview and rejects AC in TFCL test	Josephine	Uses scientific causality in interview & rejects AC in TFCL test	None
Uses the AC in interviews and in TFCL test (or guesses)	Anna	Uses scientific causality and no AC in interview. Rejects AC in TFCL test (albeit not confidently for Shirley)	Explains how used to help explanation when asked
	Beth		Explains how used to help explanation when asked
	Ramla		None
	Syeda		Used to help explanation
	Shirley		Used to help explanation
AC used in TFCL test only	Hannah	Uses scientific causality and no AC in interview. Rejects AC in TFCL test	Explains how used to help explanation when asked
	Mary		Used to help explanation
	Karen		Explains how used to help explanation when asked
	Helen		Used to help explanation
Mixed response in TFCL Test. AC not used in interview.	Ruth	Uses AC in interview but rejects in TFCL test	None
Uses AC in interview and TFCL test	Mariam	Uses AC in interview but rejects in TFCL test	None
Uses AC in interview and guesses in TFCL test	Louise	Uses AC in interview and varied responses in TFCL test	None
	Ahmed		Uses to support AC when asked about activities

Table

Analogical model = ‘Solar system’ model	Target = Atom
Shared attributes
the sun has most of the mass	the nucleus has most of the mass
the sun is the centre	the nucleus is in the centre
planets are smaller than the sun	the electrons are smaller than the nucleus
planets orbit the sun	electrons orbit the nucleus
planets and the sun attract each other	electrons and nucleus attract each other
solar system is mostly space	atom is mostly space
Unshared attributes
planets differ in size	electrons are all the same size
one planet per orbit	multiple electrons per level
planets in elliptical orbits	electrons do not orbit like planets
sun-planet force is gravity	electron-nucleus force is electrostatic

Table

Analogy = Bonfire	Target = Attraction between nucleus and an electron
Shared attributes
Unshared attributes